John Thorn makes a solid point

Post Reply
User avatar
oldecapecod11
Posts: 1054
Joined: Sat Oct 11, 2014 8:45 am
Location: Cape Haze, Florida

John Thorn makes a solid point

Post by oldecapecod11 »

ARCHIVE

N.B. Entire Archive was too large for 1 Post so it is continued in 1st Reply.

John Thorn makes a solid point
Started by John Turney, Feb 08 2014 03:25 PM

Page 1 of 3

58 replies to this topic

#1 John Turney
Pro Bowler
PFRA Member
1,522 posts
Gender:Male
Posted 08 February 2014 - 03:25 PM
“For a whole generation of fans and fantasy players,” Thorn writes, “stats have begun to outstrip story and that seems to me a sad thing. Even the unverifiable hogwash that passed for fact or informed opinion in baseball circles not so long ago seems today wistfully enticing, for its energy if nothing else.”

#2 Reaser
Pro Bowler
PFRA Member1,729 posts
Gender:Male
Location:WA
Posted 08 February 2014 - 03:49 PM
There's a few people here that try to push "on the field" / film / watching football over stats, numbers and reading box scores.

Though that isn't exactly what he was saying - plus it's an entirely different sport, one where stats actually can tell a larger part of the story.

Good point regardless, I don't know if dehumanizes (the non-derogatory definition) is the correct word, but it's the one that comes to mind.

edit to add: I also think it's a bit of a cop-out to blame fantasy (football) or people who enjoy stats in terms of lack of 'stories'. I'de go the other way and say it's the generation of media, which is more about tabloid, fabricating stories not of tall-tales or heroic feats but rather drugs, sex, affairs, and scandal. That's what sells today. The current climate of society plays a part too. And the sport itself is over-officiated and over-regulated, and much too corporate. There's little trust between players and media, it's a game of "got ya!", plus readers themselves are much less likely to enjoy say, a Johnny Blood type, etc ... Everyone wants to be offended by everything. They act like they want to know but they only want to know so they can be - or pretend to be - offended by it. Seems the most interesting locker room story we'll get these days is a mentally weak player playing the victim to potentially set up a lawsuit. So ya, stats are a problem in football in terms of evaluating players (single game, seasons, career), but for actual storytelling, stats aren't the issue, the issue is the media and society, in general. It's a different time, obviously.

#3 anome8
Second-Stringer
PFRA Member
58 posts
Gender:Male
Location:Spokane
Posted 08 February 2014 - 04:51 PM
I think what Thorn is saying is that focusing on statistics as Truth is taking the romance out of the game, for him at least. In Ken Burns' Baseball he makes a statement to the effect that baseball seduces the observer, and that the observer knows he or she is being seduced, but doesn't care. Thorn probably also feels a little culpable as he spent some time in the statistical vanguard a couple of decades ago. Basically, he's seen baseball stark naked and he'd like to see the game put some clothes back on and preserve a little mystery even if that involves being lied to.

That said, I think his complaint sounds a lot like the "today's music sucks, it was way better when I was a teenager" argument. There is no right way to enjoy a sport. We take from it what gives us pleasure and my choices shouldn't matter to anyone else and vice versa. Thorn seems to be waving his cane from a park bench and saying "you youngsters are doing it wrong." Well, they're not, and he should know better. The stories are there if you want stories. Statistics are there if you want to experience the game in a different way. Stats are just tools and the meaning is in how we use them.

#4 Jeffrey Miller
Veteran
PFRA Member
756 posts
Gender:Male
Location:Western New York
Interests:Buffalo Pro Football History
1920s NFL
1960s AFL
Posted 08 February 2014 - 05:41 PM
Are stats more important than what happens on the field? Do stats tell how important a certain tackle was as compared to another? Until recently, one didn't know (simply by looking at the stat sheet) if a 90-yard TD pass was the result of a bomb or a little dump pass that the receiver turned into a long TD catch-and-run? Too much emphasis is put on stats, in my estimation. Although they are important, as are wins and championships, they are not the end-all and be-all. Go back and watch the film to see what kind of hands Fred Biletnikoff had, or how a certain pass rusher was successful because he had so many methods of getting to the quarterback. Yes, we know how many passes Andre Reed and Jerry Rice caught, but do we know how many of them were spectacular? Or how many yards after the initial hit did Earl Campbell gain? No!! You must go back to the film, and in the case of someone like Benny Friedman, back to the newspaper accounts, to know the real story.

#5 Reaser
Pro Bowler
PFRA Member
1,729 posts
Gender:Male
Location:WA
Posted 08 February 2014 - 05:45 PM
anome8, on 08 Feb 2014 - 4:51 PM, said:
I think what Thorn is saying is that focusing on statistics as Truth is taking the romance out of the game, for him at least. ....
I agree with most of what you said, at least in terms of baseball it applies. I think, though could be wrong, what the original post meant was to apply the same to football? Which is completely different, obviously. Stats in baseball do tell the story, more or less. It isn't the same for football, which is much more complex.

I've seen a comment on another site that said; "Seattle's secondary wasn't that impressive, Demaryius Thomas owned them 13/118/1 and also Welker had 8 for 84."

In baseball you either got on base or you didn't, it's not entirely that simple but you can be b&w about it ... in football there's much more; from playcalling, to personnel, to formation, pre-snap, execution and so on. Even the difference between a clock and innings is a huge difference.

It's well put - for baseball - when you say "he's seen baseball stark naked and he'd like to see the game put some clothes back on and preserve a little mystery even if that involves being lied to." It's nearly opposite in football, plus football has many more layers of clothing, ha. To continue with the analogy being fully clothed with a sweatshirt and overcoat on would equal stats in football. Devolving into stats isn't getting to the core/removing layers/naked truth of the sport (football, for clarity), it's going the other way, adding a hideous and useless Christmas sweater to the mix.

Not really an apples to apples comparison between the sports. Stats have a completely different meaning in each.

In football you have to watch to get the story, in baseball you can look at the stats and for all intents and purposes get the story (won't know if there was a checked swing that turned into a double but things like that even out over the course of a 162 games/large sample size.)

Though profits are important, the sport must take precedence over the business

#6 Jeffrey Miller
Veteran
PFRA Member
756 posts
Gender:Male
Location:Western New York
Interests:Buffalo Pro Football History
1920s NFL
1960s AFL
Posted 08 February 2014 - 05:46 PM
Good points Reaser. The two sports do not equate ..

#7 Ben
Starter
PFRA Member
494 posts
Gender:Male
Posted 08 February 2014 - 08:03 PM
There is no conflict in my mind between statistical analysis and the romance or magic of the game, and there's plenty of both to go around.

#8 TouchdownTimmy
Starter
Forum Visitors
326 posts
Gender:Male
Location:Maryland
Interests:Almost anything to do with football and sports on the field. I am not big into celebrity or social media. I just like watching and talking about sports. Football is my game and I have been following since I was five back in 1971.
Posted 08 February 2014 - 11:39 PM
In his first autobiography "Off My Chest" Jim Brown laments on how it seemed with every carry he became more of a number than a man. That was written in 1964 when there were very few stats used in football. Brown also felt like awards were based more on reputation than performance. I agree with him on both counts.

I'm not here to argue whether the game is more fun with statistics or not. It's just looked at differently. I don't play fantasy myself, because I believe that football is the ultimate team game and offensive line play should be accounted for somehow. Also, I could never root for the opposition's players if they are playing my favorite team. But for those who play, to each his own.

I like the game of football. And I like the fact that no matter who is playing and where the score is always 0-0 when they kick off which means nothing that happened beforehand matters one bit.

To me, statistics are more noise to be turned down. For me, the play is the thing.

#9 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 09 February 2014 - 09:58 AM
Ben, on 08 Feb 2014 - 8:03 PM, said:
There is no conflict in my mind between statistical analysis and the romance or magic of the game, and there's plenty of both to go around.
Absolutely.

#10 Rupert Patrick
Pro Bowler
PFRA Member
2,441 posts
Gender:Male
Location:Upstate SC
Posted 09 February 2014 - 03:52 PM
Ben, on 08 Feb 2014 - 8:03 PM, said:
There is no conflict in my mind between statistical analysis and the romance or magic of the game, and there's plenty of both to go around.

The numbers are important, but it's the stories that reside inside the numbers. Both are essential.

#11 John Turney
Pro Bowler
PFRA Member
1,522 posts
Gender:Male
Posted 09 February 2014 - 09:58 PM
Football is art. I like the stats and I love to count things, but I love the art more.

#12 Jeffrey Miller
Veteran
PFRA Member
756 posts
Gender:Male
Location:Western New York
Interests:Buffalo Pro Football History
1920s NFL
1960s AFL
Posted 09 February 2014 - 11:43 PM
John: Great way to put it ...

#13 byron
Starter
PFRA Member
259 posts
Gender:Male
Posted 10 February 2014 - 01:15 PM
John Turney, on 09 Feb 2014 - 9:58 PM, said:
Football is art. I like the stats and I love to count things, but I love the art more.

And the cheerleaders. Baseball doesn't have those either. Dang that dry, unloving, stat-focused sport!

#14 bachslunch
Veteran
Forum Visitors
613 posts
Posted 10 February 2014 - 04:28 PM
Rupert Patrick, on 09 Feb 2014 - 3:52 PM, said:

The numbers are important, but it's the stories that reside inside the numbers. Both are essential.

Agreed both are important, and I'm very sabermetric sympathetic. The way I look at it: numbers are meaningful when used correctly, but context is really important -- things like period adjustment and controlling for certain issues matter even in baseball stats. A football example would be the yahoos who have posted at places like Bleacher complaining that Jan Stenerud is a bad HoF choice while saying "just look at his FG percentage -- he wouldn't even crack a top 20 list today with a 66.8% success rate" -- ignoring completely that period-adjusted, he's one of the best all-time.

When replying on HoF worth for baseball position players, I list OPS+ and WAR but control for position played and similar number of lifetime plate appearances. When you don't, you can end up with silly extreme-case scenario arguments like "John Paciorek might have a HoF case since he had an OPS+ of 495" when his career lasted for one glorious game and 5 plate appearances at the end of 1963. Or a BBHoF with no catchers if you're comparing them with first basemen. A less extreme but still ridiculous example in football are the occasional trolling entreaties seen on behalf of Warren Wells and the PFHoF -- again, he's got way too short a career even for a Gale Sayers type exception.

Of course baseball lends itself very well to sabermetric thinking, while that's trickier for football.

#15 Rupert Patrick
Pro Bowler
PFRA Member
2,441 posts
Gender:Male
Location:Upstate SC
Posted 10 February 2014 - 06:14 PM
bachslunch, on 10 Feb 2014 - 4:28 PM, said:

Agreed both are important, and I'm very sabermetric sympathetic. The way I look at it: numbers are meaningful when used correctly, but context is really important -- things like period adjustment and controlling for certain issues matter even in baseball stats. A football example would be the yahoos who have posted at places like Bleacher complaining that Jan Stenerud is a bad HoF choice while saying "just look at his FG percentage -- he wouldn't even crack a top 20 list today with a 66.8% success rate" -- ignoring completely that period-adjusted, he's one of the best all-time.

When replying on HoF worth for baseball position players, I list OPS+ and WAR but control for position played and similar number of lifetime plate appearances. When you don't, you can end up with silly extreme-case scenario arguments like "John Paciorek might have a HoF case since he had an OPS+ of 495" when his career lasted for one glorious game and 5 plate appearances at the end of 1963. Or a BBHoF with no catchers if you're comparing them with first basemen. A less extreme but still ridiculous example in football are the occasional trolling entreaties seen on behalf of Warren Wells and the PFHoF -- again, he's got way too short a career even for a Gale Sayers type exception.

Of course baseball lends itself very well to sabermetric thinking, while that's trickier for football.

I think you cannot have one without the other, there is a place for stats, but there is also a place to document and study the great games and players and rich history. No amount of analysis can determine which QB had the tightest spiral, or which running back could change direction the best.

Football is difficult to analyze; I have no idea how to rate linemen, and how you balance the contributions of a cornerback against the contributions of a punter.

#16 TouchdownTimmy
Starter
Forum Visitors
326 posts
Gender:Male
Location:Maryland
Interests:Almost anything to do with football and sports on the field. I am not big into celebrity or social media. I just like watching and talking about sports. Football is my game and I have been following since I was five back in 1971.
Posted 10 February 2014 - 09:45 PM
byron, on 10 Feb 2014 - 1:15 PM, said:

And the cheerleaders. Baseball doesn't have those either. Dang that dry, unloving, stat-focused sport!
Neither do the Steelers. One more reason why they are my favorite team.

#17 luckyshow
Starter
Forum Visitors
360 posts
Posted 11 February 2014 - 07:08 AM
byron, on 10 Feb 2014 - 1:15 PM, said:

And the cheerleaders. Baseball doesn't have those either. Dang that dry, unloving, stat-focused sport!
There are cheerleaders in the Korean major league (KBO), they have short skirts and pom poms and do their thing mostly from the top of the dugouts.


#18 luckyshow
Starter
Forum Visitors
360 posts
Posted 11 February 2014 - 07:19 AM
TouchdownTimmy, on 10 Feb 2014 - 9:45 PM, said:
Neither do the Steelers. One more reason why they are my favorite team.
The ridiculous "competition" by fans as to this sport is better than that sport, is pointless. And since NFL is most popular by far, it seems kind of odd, like being unsure and defensive, to dis baseball from high above.

Is football still the "scientific" sport? Is that what the claim is? There was no romance in the Gashouse Gang and Dizzy and Daffy, or Murderers Row, or even of just the "Green Monster" or just the extant existance of Wrigley Field?

Conversely is there romance to the bastardization of Soldier Field?

OK, off the subject, sure. I usually am...

I'm just saying. I think the is coming down on the obsession of stats, not for stats sake or comparisons aas much as FAntasy football. Maybe it is gambling that is the culprit. A partial reason for all the annoying graphics constantly on the screen.as well as rooting for unecessary plays at games' end to bolster fantasy roster layers' stats. And for those not working in sports talk radio who need to see every game with Sunday Ticket. Without gambling, I presume the NFL would suffer in a lower viewership and fan base, no? And seems a reason that drugs in football should be as punished as it is in baseball, when it seems the opposite way, the opposite reastion. It seems they suspend players in NFL more for pot than for the enhanceent stuff.

.
#19 coach tj troup
Veteran
PFRA Member
585 posts
Gender:Male
Posted 11 February 2014 - 03:10 PM
....to paul cesane(a sabol favorite)"all art is selected detail"....can stats and evaluation of players co-exist? have tried to find correlations between football & baseball with limited success. best example of differences? first favorite pitcher was warren spahn(cubs did not have pitchers in those days....they had guys who went to the mound to be shelled) I respond to dad what a great pitcher spahn is.....my dad's response is "if he is so damn good, how come he doesn't pitch against the dodgers in brooklyn"...this is 1960, so at the age of nine, am lost. look it up at baseball reference....he starts 1 game in '52, and 1 game in '53 in brooklyn, and gets pounded in both....his w/l for 53-55 61 wins and 33 losses. so he can beat the rest, just not the dodgers on the road, nor does he get the chance. len ford is unblock able in early 50's....but not so much against nyg....young rosy brown does not take the day off, he learns on the job how to block this monster. marchetti the same for colts, yet am sure lombardi was all over gregg leading up to the game that week. some correlations can be made, some cannot. as someone who relishes the stats, but loves film even more....can't we have both?


#20 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 11 February 2014 - 04:05 PM
Let's take a step back and ask the bigger question: what constitutes statistical analysis?

Sadly, most of what passes for statistical analysis these days is back fitting and backwards looking, e.g., the fastest car won the race, with voluminous stats illustrating how the car with the best time finished first. To put this observation in layman's terms: duh.

Little attempt, though, to dig deeper and understand the why and the how. Today's fragmented media knock each other over to be the first to information that anyone can look up.

The future of sports analysis will focus on what will happen tomorrow, not gathering stats to support what happened yesterday. Yesterday's game--no matter how well articulated and statisticaly supported--will always be burdened with subjectivity.

But...find the stat that predicts what happens tomorrow and you will have invented the proverbial better mousetrap. The world will beat a path to your door.

Page 1 of 3

oldecapecod 11

John Thorn makes a solid point
Started by John Turney, Feb 08 2014 03:25 PM

Page 2 of 3

58 replies to this topic

#21 John Turney
Pro Bowler
PFRA Member
1,522 posts
Gender:Male
Posted 11 February 2014 - 05:30 PM
coach tj troup, on 11 Feb 2014 - 3:10 PM, said:
as someone who relishes the stats, but loves film even more....can't we have both?
I think so. I think both are good.

#22 Jeffrey Miller
Veteran
PFRA Member
756 posts
Gender:Male
Location:Western New York
Interests:Buffalo Pro Football History
1920s NFL
1960s AFL
Posted 11 February 2014 - 07:51 PM
Timmy Smith gained 200 yards in a Super Bowl.

#23 Bernard Brinker
Starter
Forum Visitors
152 posts
Gender:Male
Posted 11 February 2014 - 11:17 PM
Veeshik_ya, on 11 Feb 2014 - 4:05 PM, said:
Let's take a step back and ask the bigger question: what constitutes statistical analysis?

Sadly, most of what passes for statistical analysis these days is back fitting and backwards looking, e.g., the fastest car won the race, with voluminous stats illustrating how the car with the best time finished first. To put this observation in layman's terms: duh.

Little attempt, though, to dig deeper and understand the why and the how. Today's fragmented media knock each other over to be the first to information that anyone can look up.

The future of sports analysis will focus on what will happen tomorrow, not gathering stats to support what happened yesterday. Yesterday's game--no matter how well articulated and statisticaly supported--will always be burdened with subjectivity.

But...find the stat that predicts what happens tomorrow and you will have invented the proverbial better mousetrap. The world will beat a path to your door.

I come at this from a different perspective. I am more than a little wary of statisticans or soothsayers who peer into the future of complex systems. Not sure who said it but a line that has stuck with me: "only fools, liars, or charlatans predict the future."

#24 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 11 February 2014 - 11:52 PM
Bernard Brinker, on 11 Feb 2014 - 11:17 PM, said:

I come at this from a different perspective. I am more than a little wary of statisticans or soothsayers who peer into the future of complex systems. Not sure who said it but a line that has stuck with me: "only fools, liars, or charlatans predict the future."

Then how do you explain a guy like Nate Silver, who developed a system for forecasting the performance of MLB players, and who correctly predicted the winner of all 50 states in the last presidential election?

That the sea is full of snake oil salesmen is no reason to discount legitimate stat based forecasting. It's the future.

#25 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 11 February 2014 - 11:55 PM
Btw, John Thorn is right.

If Shakespeare were alive today, he'd throw fantasy players and (most) stat geeks in with the lawyers.

#26 Reaser
Pro Bowler
PFRA Member1,729 posts
Gender:Male
Location:WA
Posted 12 February 2014 - 02:20 AM
Besides that it was a baseball , which is a sport where stats matter much more (understatement) which I think has led to some confusion ... I think "we" - and I'm guilty of this - are using the word "stats" too generally in this thread, in terms of the negativity. It probably is more about how people use stats;

"Fred Taylor has more rushing yards, more 1,000 yard seasons, better yards per carry, more TD's, more yards per game, more receiving yards, and so on than Larry Csonka, so Fred Taylor is better." I've never seen anyone actually say that but we've all seen similar examples, thankfully happens less on these forums than other places, but still happens.

Or things like, I want to say NWebster posted it a week or two ago? about how during HOF voting the presentation on Floyd Little said he was hit in the backfield whatever percentage of the time, and he wondered compared to what? and I agree, I see that "stat" and I think so what? Compared to who and what? What the hell does that even show or mean? Stats with no context.

We ask "is player A HOF worthy?" and instead of talking about what type of player he was, the first thing said is "he has x amount of (meaningless stat)" ...

and people don't understand eras.

Then there's things like "Bobby Layne wasn't even good, look how many interceptions he threw" ... said by people who've never watched any film of him and are just going off some stats and don't talk about things that actually matter (side note: infuriating that people always leave out the play-calling aspect) ...

Then you have Elias, who hands Manning the passing yardage record, which to me is a huge blow to the integrity of their statkeeping. That's not even getting into numerous errors that people try to correct and it seems hit and miss on whether they choose correct it or not. So the numbers being used aren't even correct in some cases and the people keeping the numbers evidently don't care.

Plus of course the usage of "since the merger" stats, "he has the most (whatever) since the merger" ... okay?
Then there's the fact for the first decade plus there wasn't even (official) stats, and other statisical categories took even longer to become official and/or became a stat. So can't compare. Then there's the sack numbers, which people have gone back and done A LOT of work but can only go back so far, and there's still what's "official" which leads to some idiotic posts about pass rushing on various forums.

Defensive stats in general, tackles? Ha! Some of the ridiculously inflated tackle numbers we see, some guys get credited just for being within 10 yards of the play. Then there's the type of tackles, "LB'er 'A' got x amount of tackles!" Then the people that actually watched the game see he made them all 10-20 yards down the field and was credited with 3 more tackles than he actually had, while LB'er "B" got credited with a few less tackles but they all came at the LOS or for 1-2 yd gains. Those things don't show up in "stats" ... and definitely not in the stats people like to use, in general.

Interceptions are one of my favorite things in football (majority of my favorite players are DB's) ... I love seeing interceptions. There's actual - obviously - correlations between turnovers and winning so it can be a useful stat ... but the interceptions themselves, see all the time people not knowing what they're talking about. A shitty CB who gets thrown at all season gets x amount of interceptions and people say he had a good year because? "He had (however many) interceptions!" ... A CB shuts everything down, never gets thrown at and gets 0-2 interceptions and media and fans are wondering what's wrong with him. Makes no sense. (to be fair, this was more a problem prior to more recent times, people finally caught/catching up on it.)

Even kicking and punting stats, doesn't take into account field, weather, in some cases people just blow right past caring about distance even.

Nearly half the offense doesn't even have individual stats (linemen) yet stats are so 'important' for the other 6 players on that side of the ball? TE's get broken down to simple rec/yds/TD, no consideration for blocking, where's the WR stat for WR's that actually know how to block? D-linemen get broken down to just sacks, it's lazy. Pass rushing OLB'ers get honors while OLB who do more than just rush the passer don't get talked about. And so on.

It's a rambling post and I'm throwing it all out there, but those are the types of issues with stats. It's my own personal opinion but I don't think you can "know" football without actually watching football, of course next step is knowing what you're watching, and I also believe it helps to have played, more years and higher level the better but really even half a season of JV Jr. High football might be enough, at least the exposure, having the perception of being on the field and knowing what it "looks" like from being in it, I think that matters as well.

Either way, I still enjoy stats as much as anyone, still have my thousands of football cards, was always look at the picture then flip it to the back and look at the stats. Hasn't been a day in I can't even remember how long where I didn't look up or see football related stats. I think the point though is that when you get into discussions about football, real football which is on the field, the discussion too often gets corrupted by the injection of useless numbers and/or stats lacking context, or all of the above.

My thoughts, at least.

Though profits are important, the sport must take precedence over the business

#27 JWL
Pro Bowler
PFRA Member
1,846 posts
Gender:Male
Posted 12 February 2014 - 07:35 AM
Did Jimmy Graham get selected first team all-pro at tight end due to stats or was he really a better tight end than Vernon Davis?

#28 Bryan
Veteran
Forum Visitors
662 posts
Posted 12 February 2014 - 09:58 AM
Veeshik_ya, on 11 Feb 2014 - 4:05 PM, said:
Let's take a step back and ask the bigger question: what constitutes statistical analysis?

Sadly, most of what passes for statistical analysis these days is back fitting and backwards looking, e.g., the fastest car won the race, with voluminous stats illustrating how the car with the best time finished first. To put this observation in layman's terms: duh.

Little attempt, though, to dig deeper and understand the why and the how. Today's fragmented media knock each other over to be the first to information that anyone can look up.

The future of sports analysis will focus on what will happen tomorrow, not gathering stats to support what happened yesterday. Yesterday's game--no matter how well articulated and statisticaly supported--will always be burdened with subjectivity.

But...find the stat that predicts what happens tomorrow and you will have invented the proverbial better mousetrap. The world will beat a path to your door.

I agree with all of this. Baseball is a more stat-driven sport, but a whole field of statistics is devoted to analyzing minor league performance and then predicting how these stats would translate to the major leagues. With so much hype devoted to the NFL draft (and even college recruiting), I see the same type of statistical field developing for football (although I don't think it will be nearly as successful...too many variables in football stats). With the amount of money wagered on football games, if someone can come up with a reliable predictive model it would be huge.

The "backwards-looking" statistical analysis sometimes drives me crazy, because it can be used to completely misconstrue reality. I think it goes hand-in-hand with Reaser's complaint about a lack of context. I read a baseball article about Rick Reuschel...lots of statistical gyrations and so forth...very detailed...the gist of the article was that Reuschel played for a long time and was an above-average pitcher on a lot of lousy teams, and through some statistical manipulations we see that Rick Reuschel was/is actually a HOF pitcher. Kind of weird. I watched Rick Reuschel pitch, I followed baseball during Reuschel's career, and at no point in time did anyone think he was a HOF pitcher. Its kind of like "Rick Reuschel was a Hall of Famer...who knew?", but things get distorted through the statistical looking glass.

It would make me laugh when Mike Tanier would write some retro-statistical piece and rhetorically opine things like "Charley Taylor only had a single 1000-yard season in his career" or "Paul Warfield made the pro bowl and only caught 29 passes", as if the achievements of the historical players were somewhere between myth and fallacy, and Tanier was the Indiana Jones uncovering the Ark, showing us that Reggie Rucker was indeed superior to Lynn Swann. I enjoy looking backwards at stats, but sometimes I think the "analysis" can do more harm than good, especially if the analyst lacks historical context.

#29 JWL
Pro Bowler
PFRA Member
1,846 posts
Gender:Male
Posted 12 February 2014 - 11:39 PM
Yeah, Tanier wrote that "Warfield damage" article and it was discussed here.

I also find the Reuschel stuff to be ridiculous. He was a good pitcher when he was active. Now I see him on top 100 all-time greatest player lists.

#30 RickRamacier
Second-Stringer
PFRA Member
87 posts
Gender:Male
Location:Martinez, CA
Interests:Football History.
Posted 13 February 2014 - 01:16 AM
Broadly speaking, stats in football are important and of course interesting. But, they only tell part of the story. An over reliance on them might lead one to think the 73 Bills were very a good team. In-person observations and an understanding of those observations are very important; as is film. Good writing by knowledgeable writers helps tell the story too. Stats are just numbers. Numbers are abstract representations of things that need definition, context, and relative value.

The statement makers that suggest that the Seattle secondary did not have a strong super bowl are relying on numbers without their correct definition, context or understanding. Thus, the stats relied on take on meaningless, and the statements are baseless.

Our schools have taught the last coupe of generations to think within narrow silos that often contain a series of math equations (teaching to the test) without regard to critical thinking about them (context, meaning, definition, etc.). For many of these folks, going beyond the pure numbers is tough.

One of the main reasons the steroid era is so bothersome to avid baseball fans is because it complicates the abstract nature of the sport's numbers that ordinarily do not need as much context or definition as say - the numbers from football.

#31 bachslunch
Veteran
Forum Visitors
613 posts
Posted 13 February 2014 - 08:53 AM
RickRamacier, on 13 Feb 2014 - 01:16 AM, said:
One of the main reasons the steroid era is so bothersome to avid baseball fans is because it complicates the abstract nature of the sport's numbers that ordinarily do not need as much context or definition as say - the numbers from football.

This kind of problem still exists in other circumstances regarding baseball, though. For example, arguing in favor of a "real" or "pure" single-season home run record is a futile exercise since Roger Maris played during a time when amphetamine use was rife (and his teammate and co-record chaser that year, Mickey Mantle, got under-the-table shots containing steroids and amphetamines and plenty else for pain issues -- which incidentally backfired on him when he missed some playing time after getting an infection from one). Babe Ruth (and everyone else before him) set their single-season records pre-integration, and Ruth apparently tried using goat testosterone at one point during his career (stopping because he got sick from it). And just as importantly, both Ruth and Maris were lefty hitters who took full advantage of Yankee Stadium's short RF porch -- in fact, those dimensions were created with Ruth in mind.

#32 Bob Gill
Pro Bowler
PFRA Member
1,200 posts
Gender:Male
Posted 13 February 2014 - 09:16 AM
Just for the record, Babe Ruth hit 20 more home runs on the road for his career than he did at home. And Roger Maris in 1961 hit 30 at home, 31 on the road.

#33 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 13 February 2014 - 11:16 AM
I have agreed with just about everything said in this thread, on both sides of the topic. I'd like to add a couple of small comments.

Certainly the romance and the legends are a big part of the game, and they should be. But, as researchers and historians, aren't we obligated to look into some of these legends to see how much truth has gone into them? And doesn't that mean sometimes debunking a legend, however much it adds to the romance?

While honoring the heroes and the legends of the past, don't we all wish we had more stats from the earlier years of pro football? I have heard several of Don Hutson's teammates say that he never dropped a catchable pass. I have heard contemporaries say the same about Raymond Berry. I'm sure neither statement is quite true, but how much truth do they contains? Wouldn't it be great to know how many times they were targeted and how many passes they did drop.

And wouldn't it be even greater to have complete stats for, e.g., Red Grange and Ernie Nevers and Benny Friedman and Verne Lewellen and hundreds of others from before 1932?

#34 Bob Gill
Pro Bowler
PFRA Member
1,200 posts
Gender:Male
Posted 13 February 2014 - 11:35 AM
Absolutely.

A good case regarding Hutson is the long-held belief that he caught passes in something like 95 straight games. But in the Pro Football Chronicle they made a very convincing case that he had no catches in one game about in the middle of the streak. Historians are supposed to be on the lookout for things like that. In the same way, if we find that some famous old story turns out to be demonstrably false, we should point that out too. The game's "flavor" or "color" should still be based on the truth.

#35 byron
Starter
PFRA Member
259 posts
Gender:Male
Posted 13 February 2014 - 12:41 PM

While honoring the heroes and the legends of the past, don't we all wish we had more stats from the earlier years of pro football?

It would be great to have complete stats for the earlier years of Pro Football. I find the stats get pretty sparse starting the early-to-mid 40s which leaves a lot of room for research into the early decades of the game.

The other part of this is that we now live, as has been pointed out, in a stat-driven sports society. Based on the sports society of the early decades of Pro Football, the stats we have are complete. We just want more, although it's worth pointing out that we don't need more to enjoy the games from those early eras.

#36 Bernard Brinker
Starter
Forum Visitors
152 posts
Gender:Male
Posted 13 February 2014 - 02:07 PM
Veeshik_ya, on 11 Feb 2014 - 11:52 PM, said:
Then how do you explain a guy like Nate Silver, who developed a system for forecasting the performance of MLB players, and who correctly predicted the winner of all 50 states in the last presidential election?

That the sea is full of snake oil salesmen is no reason to discount legitimate stat based forecasting. It's the future.

"It's the future"---I would add: and it always will be.

I understand the need for prediction in economic matters or on earthquakes. Unfortunately, according to Nate Silver, such systems all but defy accurate prediction unlike baseball and presidental elections.

I am not sure why fans of baseball, or presidental elections for the matter, need forecastes (accurate or not) of the future. Why not sit back and enjoy the show. Ohh, I almost forgot: fantasy. Then again Hari Seldon is also fantasy.

#37 bachslunch
Veteran
Forum Visitors
613 posts
Posted 13 February 2014 - 04:12 PM
Bernard Brinker, on 13 Feb 2014 - 2:07 PM, said:

I am not sure why fans of baseball, or presidental elections for the matter, need forecastes (accurate or not) of the future. Why not sit back and enjoy the show. Ohh, I almost forgot: fantasy. Then again Hari Seldon is also fantasy.

No question that MLB front offices would love to have accurate forecasts of baseball players' futures -- they want to sink free agent money into the surest things possible. As for fans, forecasts a layman can understand would make discussions about players more informed, and not just going forward. Much of the "who should be in the HoF" chatter one reads tries to make backwards sense of players, especially more borderline cases.

#38 byron
Starter
PFRA Member
259 posts
Gender:Male
Posted 13 February 2014 - 04:37 PM

Stat based prediction is the future---I would add: and it always will be.

In a dry, ugly world--yes. In a world of story and excitement, where one enjoys the game, not the analysis, it will never take root.

#39 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 13 February 2014 - 05:43 PM
Veeshik_ya, on 11 Feb 2014 - 11:55 PM, said:
Btw, John Thorn is right.

If Shakespeare were alive today, he'd throw fantasy players and (most) stat geeks in with the lawyers.But
But Shakespeare didn't say "Let's kill all the lawyers." Dick the Butcher said it and he was a follower of Jack Cade, a character with whom Shakespeare cleaerly did not sympathize

#40 Bernard Brinker
Starter
Forum Visitors
152 posts
Gender:Male
Posted 13 February 2014 - 10:16 PM
byron, on 13 Feb 2014 - 4:37 PM, said:

In a dry, ugly world--yes. In a world of story and excitement, where one enjoys the game, not the analysis, it will never take root.

Sorry Byron my sentence was not very clear---I have edited it. I was attempting to get across that such accurate forecasting is not in the cards, it will always be in our future, out of our current grasp.

Page 2 of 3

N.B. Too much text so continued in 1st Reply

oldecapecod 11
"It was a different game when I played.
When a player made a good play, he didn't jump up and down.
Those kinds of plays were expected."
~ Arnie Weinmeister
User avatar
oldecapecod11
Posts: 1054
Joined: Sat Oct 11, 2014 8:45 am
Location: Cape Haze, Florida

Re: John Thorn makes a solid point

Post by oldecapecod11 »

ARCHIVE

John Thorn makes a solid point
Started by John Turney, Feb 08 2014 03:25 PM

N.B. Too much text for 1 Post so continued here in 1st Reply
oldecapecod 11

Page 3 of 3

58 replies to this topic

#41 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 14 February 2014 - 12:04 AM
Bernard Brinker, on 13 Feb 2014 - 10:16 PM, said:
Sorry Byron my sentence was not very clear---I have edited it. I was attempting to get across that such accurate forecasting is not in the cards, it will always be in our future, out of our current grasp.

Again: then how do you explain a guy like Nate Silver, who has clearly had success forecasting sports performance and politics?

Generally, I share your skepticism. But to not acknowledge legitimate succcesses like Silver is burying your head in the sand.

No snark intended here, but you sound like a person 100 years ago who refused to admit that man would ever invent mechanical machines that would fly through the sky.

#42 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 14 February 2014 - 12:50 PM
Veeshik_ya, on 14 Feb 2014 - 12:04 AM, said:
Again: then how do you explain a guy like Nate Silver, who has clearly had success forecasting sports performance and politics?
Generally, I share your skepticism. But to not acknowledge legitimate succcesses like Silver is burying your head in the sand.

No snark intended here, but you sound like a person 100 years ago who refused to admit that man would ever invent mechanical machines that would fly through the sky.
Silver's work, which I have followed rather closely, really isn't relevant, because he doesn't work with past data, he uses current data from polls and he adjusts the results for known biases and built-in error. He didn't predict the 2012 election based on 2008 results, he predicted it based on surveys taken a day or two before and he adjusted his results daily, based on the very latest surveys.

Baseball is more predictable than football because, as any bettor knows, you bet on the starting pitchers. If a very good pitcher for a bad team is facing a poor pitcher for a good, you bet on the bad team every team. But even that method falls short, since good pitchers sometimes have bad games and bad pitchers sometimes have good games. That aspect is completely unpredictable.

I think 70% is about as good as any statistical prediction method will ever get, and I can do about that well without using any math at all. There are simply too many unmeasurable factors that go into it.

#43 Jeremy Crowhurst
Starter
PFRA Member
323 posts
Gender:Not Telling
Posted 14 February 2014 - 01:19 PM
Veeshik_ya, on 14 Feb 2014 - 12:04 AM, said:

Again: then how do you explain a guy like Nate Silver, who has clearly had success forecasting sports performance and politics?

Generally, I share your skepticism. But to not acknowledge legitimate succcesses like Silver is burying your head in the sand.

No snark intended here, but you sound like a person 100 years ago who refused to admit that man would ever invent mechanical machines that would fly through the sky.
To add to what rhickok said, baseball gives you a larger data set to work with than football, and baseball is the least team-oriented of the team sports. You often can't tell how much of a run's success is dependent on the RB or the blocking. But when a fielder drops a pop fly in the outfield, there's no question where the responsibility lies.

Silver's book is a great read. But as far as football and it's very limited sample size goes, to borrow from his title, if a WR has a 1,000 yard season, then the following summer he has two DUI arrests, I'm inclined to think that the 1,000 yards are the noise and the DUIs are the signal, not the other way around.

#44 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 14 February 2014 - 03:31 PM
Silver's work is relevant, insofar as he's using statistics to predict future events. Nowhere in my posts did I attempt to pinpoint which statistics will be used or not used.

Point is: the future of sports analytics will involve looking ahead, not back. Today's media largely take existing stats after the fact and mold them to prove what just happened.

I guess we'll let time be the judge.

#45 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 14 February 2014 - 03:36 PM

Regarding the comment that any bettor knows that you bet on the starting pitcher in baseball, maybe you're not familiar with the pioneering work of Jim Barnes, who said:

"For more than three decades I have insisted that baseball is "80-percent batting and 20-percent pitching." I knew the idea had merit from the beginning and a year before leaving the Stardust to return to his native state, sportsbook manager Walker said he was coming around to my point of view. There are only a handful of pitchers that can win when their team is not scoring and the other 95-percent have their records determined by whether the team scores or not."

#46 bachslunch
Veteran
Forum Visitors
613 posts
Posted 14 February 2014 - 04:47 PM
Bryan, on 12 Feb 2014 - 09:58 AM, said:
The "backwards-looking" statistical analysis sometimes drives me crazy, because it can be used to completely misconstrue reality. I think it goes hand-in-hand with Reaser's complaint about a lack of context. I read a baseball article about Rick Reuschel...lots of statistical gyrations and so forth...very detailed...the gist of the article was that Reuschel played for a long time and was an above-average pitcher on a lot of lousy teams, and through some statistical manipulations we see that Rick Reuschel was/is actually a HOF pitcher. Kind of weird. I watched Rick Reuschel pitch, I followed baseball during Reuschel's career, and at no point in time did anyone think he was a HOF pitcher. Its kind of like "Rick Reuschel was a Hall of Famer...who knew?", but things get distorted through the statistical looking glass.

I'd contend that this tells us something about Reuschel we might not have guessed. Reuschel's pitching stats are at the the very least borderline HoF worthy -- and this is the problem as I see it with the "eye test." Reuschel did not have overpowering stuff, didn't exactly cut a showy persona (was kind of a big blobby guy), and pitched well on some bad teams that gave him little run support. Bert Blyleven is actually the poster boy for the type of pitcher who was a heck of a lot better then people remember, to the point where he's a no-doubt HoF-er and rightly in.

Reuschel pitched 3548.1 innings, putting up 70 WAR and a lifetime ERA+ of 114. For some context, here are some HoF pitchers with similar innings pitched:

Jim Bunning: 3760.1 IP/59.5 WAR/115 ERA+
Joe McGinnity: 3441.1/57.8/120
Juan Marichal: 3507.0/63.1/123
Bob Feller: 3827.0/63.6/122
Waite Hoyt: 3762.1/51.8/112

Reuschel doesn't look out of place here, actually. He has more WAR than anyone here, though his ERA+ is only better then Hoyt's.

The opposite is the HoF case for Jack Morris. He had the swagger, the big fastball, a World Series Game Seven for the ages, and a pile of wins despite not pitching at elite level. Had he been elected, he'd be one of the worst pitchers in the HoF, along with Catfish Hunter, Rube Marquard, and Herb Pennock. Morris had 44.1 career WAR and an ERA+ of 105 in 3824.0 innings, which is far below all the pitchers in the Reuschel comparison group.

Myself, I'm very suspicious of "eye test" arguments without the numbers to corroborate it.

#47 Jeremy Crowhurst
Starter
PFRA Member
323 posts
Gender:Not Telling
Posted 14 February 2014 - 04:47 PM
Veeshik_ya, on 14 Feb 2014 - 3:36 PM, said:
Regarding the comment that any bettor knows that you bet on the starting pitcher in baseball, maybe you're not familiar with the pioneering work of Jim Barnes, who said:

"For more than three decades I have insisted that baseball is "80-percent batting and 20-percent pitching." I knew the idea had merit from the beginning and a year before leaving the Stardust to return to his native state, sportsbook manager Walker said he was coming around to my point of view. There are only a handful of pitchers that can win when their team is not scoring and the other 95-percent have their records determined by whether the team scores or not."

Well... yeah... but the 80% that is batting is the same players every game. The 20% that is pitching is a different matchup most games. So when you're looking for variables to exploit, that's the first one you look at.

I'm sure Jim Barnes probably also said somewhere that it's not a good idea to bet on every single game -- that you pick your spots based on games where you think the line is off.

As for Nate Silver and football, Google "Nate Silver football" and you can read about his amazing success predicting football, like last year's Patriots - Seahawks Super Bowl....

#48 Veeshik_ya
Starter
Forum Visitors
112 posts
Gender:Male
Posted 14 February 2014 - 05:56 PM
Jeremy Crowhurst, on 14 Feb 2014 - 4:47 PM, said:
Well... yeah... but the 80% that is batting is the same players every game. The 20% that is pitching is a different matchup most games. So when you're looking for variables to exploit, that's the first one you look at.

I'm sure Jim Barnes probably also said somewhere that it's not a good idea to bet on every single game -- that you pick your spots based on games where you think the line is off.

As for Nate Silver and football, Google "Nate Silver football" and you can read about his amazing success predicting football, like last year's Patriots - Seahawks Super Bowl....

The PECOTA system he sold to Baseball Prospectus turned out pretty well, though, didn't it?

Staying on topic, the original point was this: Anyone can take a bunch of numbers and use it to build a case for their argument about events that have already happened. I think we'd all agree on that.

But we're in a new era of big data, with powerful tools to analyze and exploit it. The focus will turn to analyzing tomorrow's game and players, not yesterday's.

Why is that a bad thing? I'm surprised how uncomfortable that (apparently) makes some people.

#49 Bernard Brinker
Starter
Forum Visitors
152 posts
Gender:Male
Posted 14 February 2014 - 06:14 PM

#50 Jeremy Crowhurst
Starter
PFRA Member
323 posts
Gender:Not Telling
Posted 15 February 2014 - 12:58 AM
Veeshik_ya, on 14 Feb 2014 - 5:56 PM, said:

The PECOTA system he sold to Baseball Prospectus turned out pretty well, though, didn't it?

Staying on topic, the original point was this: Anyone can take a bunch of numbers and use it to build a case for their argument about events that have already happened. I think we'd all agree on that.

But we're in a new era of big data, with powerful tools to analyze and exploit it. The focus will turn to analyzing tomorrow's game and players, not yesterday's.

Why is that a bad thing? I'm surprised how uncomfortable that (apparently) makes some people.
I'm not sure who you're arguing with, or who you think is disagreeing with you. Projecting the future obviously involves analyzing data from the past and/or present. But that doesn't have anything to do with a lot of the discussion on this site that involves comparing players from different eras, and how much or little they contributed to the game. Both are interesting and useful exercises.

#51 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 15 February 2014 - 12:29 PM
Veeshik_ya, on 14 Feb 2014 - 3:36 PM, said:
Regarding the comment that any bettor knows that you bet on the starting pitcher in baseball, maybe you're not familiar with the pioneering work of Jim Barnes, who said:

"For more than three decades I have insisted that baseball is "80-percent batting and 20-percent pitching." I knew the idea had merit from the beginning and a year before leaving the Stardust to return to his native state, sportsbook manager Walker said he was coming around to my point of view. There are only a handful of pitchers that can win when their team is not scoring and the other 95-percent have their records determined by whether the team scores or not."

I am familiar with his work, but I didn't know he said that. In fact, I'm quite surprised he said that, since his formula for picking winners was based entirely on the ERA of the starting pitchers. IIRC, you double the pitcher's ERA for the season and subtract his ERA for the most recent three games.

#52 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 15 February 2014 - 12:50 PM
Veeshik_ya, on 14 Feb 2014 - 5:56 PM, said:

The PECOTA system he sold to Baseball Prospectus turned out pretty well, though, didn't it?

Staying on topic, the original point was this: Anyone can take a bunch of numbers and use it to build a case for their argument about events that have already happened. I think we'd all agree on that.

But we're in a new era of big data, with powerful tools to analyze and exploit it. The focus will turn to analyzing tomorrow's game and players, not yesterday's.

Why is that a bad thing? I'm surprised how uncomfortable that (apparently) makes some people.
I don't think anybody is saying it's bad. I'm certainly not. I simply think that there is a limit to what it can do. Could any system possibly have predicted a 43-8 Seattle victory in the Super Bowl or anything close to it?

The PECOTA system has worked pretty well. But it doesn't do what I thought we were talking about. So far as I know, it has never been used to predict game results, only season results, which is considerably easier. One common way of predicting a team's record is to go through the schedule a game at a time and predict which games the team will win and which they will lose. A knowledgeable football fan can do pretty well with that method. BUT --- it's likely to work because the mistakes balance out. The team in question will, for example, win two games that it "should have" lost and lose two games that it "should have" won, resulting in the right record for the season despite the fact that four predictions for individual games were wrong.

PECOTA has also done pretty well with projecting a player's stats for the season. But I'm positive it cannot be used to predict how well a player will do in a specific game, or in in the span of a week, or even over the period of a month. It might have predicted, for example, that David Ortiz would hit 30 home runs last season; but could it have predicted that he would hit 3 in April, 6 in May, 7 in June, 4 in July, and 6 in September? I doubt it very much.

In my years as a sportswriter, I dutifully predicted NFL games every week and I was always somewhere between 65% and 75% accurate for the season. My career average was right around 70%. Using a Radio Shack TRS-80 Model I in 1979, I developed a statistical analysis system for predicting games. It was 71% accurate. I refined that method and developed a couple of other methods using more advanced computers .... and they were always right about 70% of the time.

Now, I don't claim to be anything like an authority on how it should be done, but I haven't seen better results from people who know a lot more than I do about it. Football Outsiders, which has undoubtedly done more than anyone else to develop refined statistics, does about as well with its stat-based season forecasts as the typical sportswriter or knowledgeable does with the naked eyeball.

It's not discomfort .... it's skepticism.

#53 apbaball
Starter
PFRA Member
397 posts
Gender:Male
Posted 15 February 2014 - 03:34 PM

rhickok1109 said

"I have agreed with just about everything said in this thread, on both sides of the topic. I'd like to add a couple of small comments.

Certainly the romance and the legends are a big part of the game, and they should be. But, as researchers and historians, aren't we obligated to look into some of these legends to see how much truth has gone into them? And doesn't that mean sometimes debunking a legend, however much it adds to the romance?

While honoring the heroes and the legends of the past, don't we all wish we had more stats from the earlier years of pro football? I have heard several of Don Hutson's teammates say that he never dropped a catchable pass. I have heard contemporaries say the same about Raymond Berry. I'm sure neither statement is quite true, but how much truth do they contains? Wouldn't it be great to know how many times they were targeted and how many passes they did drop.

And wouldn't it be even greater to have complete stats for, e.g., Red Grange and Ernie Nevers and Benny Friedman and Verne Lewellen and hundreds of others from before 1932?"

I agree, it would be nice.

I also believe that those who say stats don;t tell the entire story are correct. In part because the nature of football. However, another reason is we do not have enough measures and the ones we currently have don't tell us enough at least for older seasons. INTs and sacks tell us such a small part of how a defensive player did. What tells us how well a defensive player was against the run and how they were in coverage? And on the other side of the ball, what stats are there for OL? Therefore, film review is needed.
Coaches and scouts have graded players for years and I would assume their ratings provide a lot of insight into a player's effectiveness. A lot of them may be somewhat subjective but they realize the importance of accurately evaluating performance. If those stats weren't valuable they wouldn't spend the time to do so.

#54 byron
Starter
PFRA Member
259 posts
Gender:Male
Posted 16 February 2014 - 10:38 PM

What tells us how well a defensive player was against the run and how they were in coverage? And on the other side of the ball, what stats are there for OL? Therefore, film review is needed.

I agree with this point entirely. One of the things I have found when studying 1940s football is that, while teams may play the same base defenses (5-3, 4-4, 6-2, etc...), they played them in different ways. For example, in 1945 the Eagles would line up in a 6-2 defense much of the time. When the offensive HB went into motion out of the backfield to a Flanker position, guess who moved out to cover him? The answer is: the Middle Guard (Nose Tackle)! I was very surprised to see this happen throughout games in both 1945 and 1948 (I haven't studied the other years in enough depth yet)--the Eagles would end up in a 5-3 because the Middle Guard would stand up and run out to the "flank" to cover the HB. This is something you just don't get from statistics; match-ups, and why they worked or didn't work. They are a huge part of the game.

#55 Bryan
Veteran
Forum Visitors
662 posts
Posted 17 February 2014 - 01:52 PM
bachslunch, on 14 Feb 2014 - 4:47 PM, said:
Reuschel pitched 3548.1 innings, putting up 70 WAR and a lifetime ERA+ of 114. For some context, here are some HoF pitchers with similar innings pitched:

Jim Bunning: 3760.1 IP/59.5 WAR/115 ERA+
Joe McGinnity: 3441.1/57.8/120
Juan Marichal: 3507.0/63.1/123
Bob Feller: 3827.0/63.6/122
Waite Hoyt: 3762.1/51.8/112

Reuschel doesn't look out of place here, actually. He has more WAR than anyone here, though his ERA+ is only better then Hoyt's.

I don't think WAR for pitchers is all that reliable, and ERA+ has its own set of quirks...but even so, I think a big difference between Reuschel and the pitchers you listed is that Reuschel's career doesn't really have a "peak". He never led the league in any major statistical category other than CGs and SHOs in 1987, he only had a couple years where he was in the Cy Young voting and only had a few all-star game appearances (FWIW), etc. So I don't think the "eye test" is all that deceiving in regards to Reuschel...I saw a pretty good pitcher and I saw him for a long time. Nothing stood out as being HOF-worthy. That jibes with his actual career, IMO.

#56 bachslunch
Veteran
Forum Visitors
613 posts
Posted 19 February 2014 - 08:18 AM
Bryan, on 17 Feb 2014 - 1:52 PM, said:

I don't think WAR for pitchers is all that reliable, and ERA+ has its own set of quirks...but even so, I think a big difference between Reuschel and the pitchers you listed is that Reuschel's career doesn't really have a "peak". He never led the league in any major statistical category other than CGs and SHOs in 1987, he only had a couple years where he was in the Cy Young voting and only had a few all-star game appearances (FWIW), etc. So I don't think the "eye test" is all that deceiving in regards to Reuschel...I saw a pretty good pitcher and I saw him for a long time. Nothing stood out as being HOF-worthy. That jibes with his actual career, IMO.

Several things:

-I think it's incorrect to say Reuschel had no peak. Here are the top five ERA+ seasons qualifying as a starter for the five pitchers mentioned:

Reuschel: 157, 157, 134,131,131.
Bunning: 149, 149, 143, 142, 133.
Hoyt: 146, 141, 137, 131, 121.
McGinnity: 169, 148, 138, 131, 116.
Marichal: 169, 167, 165, 144, 133.
Feller: 161, 154, 152, 130, 126.

The best is Marichal with three seasons above 160. McGinnity, and Feller both have an individually better season, but fall off below Reuschel as one goes down beyond that. And I don't see that Bunning or Hoyt are better than Reuschel. The point remains -- Reuschel is holding his own well here.

-the all-star game argument doesn't work well for baseball, as the squads are chosen less than halfway through each year. And Reuschel placed 3rd, 3rd, and 8th in CYA voting (1977, 1987, 1989). Note that Bunning did so only once (2nd in 1967), as did Marichal (8th in 1971), though admittedly the voting was different back in the 50s up to the mid-60s (only one winner for both leagues).

-I'd be interested to know what the problem is with WAR and ERA+, especially if one controls (as I did) for position, number of innings, and even type of pitcher (starter vs. reliever). Comparing WAR across positions might be another matter -- catchers, for one, tend not to fare that well by WAR compared to other positions. But that's not the case here.

-there's a difference between going to a lot of games and doing film-study level of analysis. I see people all the time argue online that they saw such-and-such a player and insist this is definitive and seriously meaningful. I'm not convinced of such approaches, particularly given all the distractions one can encounter in watching a game (not the least of which is how many beers one might have had) and just how good the evaluative skills of the viewer are.

Look, I'm not saying Reuschel is a no-brainer. But he's got a very reasonable second-tier HoF-er argument.

#57 rhickok1109
Pro Bowler
PFRA Member
1,282 posts
Gender:Male
Location:New Bedford, MA
Posted 19 February 2014 - 09:52 AM
byron, on 16 Feb 2014 - 10:38 PM, said:

I agree with this point entirely. One of the things I have found when studying 1940s football is that, while teams may play the same base defenses (5-3, 4-4, 6-2, etc...), they played them in different ways. For example, in 1945 the Eagles would line up in a 6-2 defense much of the time. When the offensive HB went into motion out of the backfield to a Flanker position, guess who moved out to cover him? The answer is: the Middle Guard (Nose Tackle)! I was very surprised to see this happen throughout games in both 1945 and 1948 (I haven't studied the other years in enough depth yet)--the Eagles would end up in a 5-3 because the Middle Guard would stand up and run out to the "flank" to cover the HB. This is something you just don't get from statistics; match-ups, and why they worked or didn't work. They are a huge part of the game.
How did they manage to have a middle guard in the 6-2? I know that they sometimes overshifted or undershifted, but either shift would put one of the guards over center, so you would have different players at middle guard at various times, so "middle guard" simply means either of the guards.

Of course, the 6-2 usually became the 5-3 in passing situations, so one of the guards would drop back to become the middle LB, while the other would move over to become the middle guard.

#58 bachslunch
Veteran
Forum Visitors
613 posts
Posted 19 February 2014 - 04:29 PM
Bryan, on 17 Feb 2014 - 1:52 PM, said:
I don't think WAR for pitchers is all that reliable

Just looked at a thread on Rick Reuschel over at the Baseball Think Factory's website. There appear to be at least three things helping to boost his WAR numbers which don't quantify readily via standard pitcher counting stats:

-Reuschel was a pretty good hitter for a pitcher.
-Reuschel fielded his position well.
-Reuschel spent a lot of his career not only getting minimal run support, he also pitched with some horrible defenses behind him (especially while in Chicago). So while (for example) Jim Palmer enjoyed having players like Mark Belanger and Paul Blair in the field behind him, Reuschel had to make do with stone-glove guys like Bill Madlock and Rick Monday backing him up.

#59 byron
Starter
PFRA Member
259 posts
Gender:Male
Posted 19 February 2014 - 06:45 PM

How did they manage to have a middle guard in the 6-2? I know that they sometimes overshifted or undershifted, but either shift would put one of the guards over center, so you would have different players at middle guard at various times, so "middle guard" simply means either of the guards.

Of course, the 6-2 usually became the 5-3 in passing situations, so one of the guards would drop back to become the middle LB, while the other would move over to become the middle guard.

I'm referring to the "Middle Guard" as whoever lined up directly in front of the offensive Center (the overshift you spoke of, perhaps). In the 40s, the MG did not drop into the MLB spot (it may have happened rarely) and cover the middle of the field. What I was referring to was that the Middle Guard actually ran outside of the DE to cover the HB when he went in motion (becoming a "Flanker")--often that motion would take the HB several yards past the DE on that side. It was not a short run.

If they were going to be efficient, the MG could have dropped into the LB spot and the LB on that side could cover the Flanker. But that wasn't happening.

Page 3 of 3
"It was a different game when I played.
When a player made a good play, he didn't jump up and down.
Those kinds of plays were expected."
~ Arnie Weinmeister
Post Reply