Perfect Ted Williams... what gives?

[ncap99]

I will also add that it has been a multi year issue with how contact works. It is like an arms race that keeps resetting the baseline. At this point, if you aren't running a high BABIP (i.e. low power) high contact card with high avoidK they will be lucky to hit .200. No amount of power within reason will offset that. Power is almost detrimental at this point, because it comes directly at the cost of getting on base (see above BABIP formula).

Take Babe Ruth as an example. His ratings are insane - 124 contact and 219 power. AvoidK on the low side at 91, but 116 eye. Looking at PeL where you'll find the highest concentration of Babe Ruth's, these are the BAs in full years on record with teams that have him:
.214 (-0.1 WAR in 573 PA)
.187 (-0.2 WAR in 333 PA)
.215 (.9 WAR in 388 PA)

These seasons were all at PeL running against teams that know how to use park factors and have the best pitching, etc.

He is sort of the poster child of how power sucks. He still hits a ton of HRs and he has slightly below average (for diamond+) defense. He also has the highest contact rating of any card in the game (at least ones in the database) and still can't hit over .220.

[frankbama]

Originally Posted by chazzycat

It's much smaller, more like 3%.


Yep the game has always been harsh on HR-hitting sluggers who also had great contact skills & batting averages. Those types of players are just so elite & so rare (Ruth, Williams, Bonds) that I don't think the algorithm for translating stats to ratings handles them very well. Across baseball history, sluggers tend to strike out too much and have lower averages, which I believe is why the system is set up that way. It handles the majority of players pretty well. But in these extreme cases it just falls apart, and of course these "extreme" cases are the players we love the most.

This year actually seems a bit better than previous, for what it's worth. Ted and Ruth are both getting significant play in the highest levels despite these perceived shortcomings.

Was not aware there was a penalty for DH so learn something every day. Is there a reason for that? In "real life" as a basis some hit better as a DH without having to field and others have said they hate it since playing the field keeps them in the game and not thinking about hitting as much. What does the game infer about this and its effects?

[frankbama]

Originally Posted by ncap99

I will also add that it has been a multi year issue with how contact works. It is like an arms race that keeps resetting the baseline. At this point, if you aren't running a high BABIP (i.e. low power) high contact card with high avoidK they will be lucky to hit .200. No amount of power within reason will offset that. Power is almost detrimental at this point, because it comes directly at the cost of getting on base (see above BABIP formula).

Take Babe Ruth as an example. His ratings are insane - 124 contact and 219 power. AvoidK on the low side at 91, but 116 eye. Looking at PeL where you'll find the highest concentration of Babe Ruth's, these are the BAs in full years on record with teams that have him:
.214 (-0.1 WAR in 573 PA)
.187 (-0.2 WAR in 333 PA)
.215 (.9 WAR in 388 PA)

These seasons were all at PeL running against teams that know how to use park factors and have the best pitching, etc.

He is sort of the poster child of how power sucks. He still hits a ton of HRs and he has slightly below average (for diamond+) defense. He also has the highest contact rating of any card in the game (at least ones in the database) and still can't hit over .220.

I think a lot of this is why it's so frustrating. The points of any game are to win and to have fun. The higher you get in the game, ironically the less realistic and fun it seems to be. I play and I still have fun with it but would like to see some ways to win without running unrealistic ballparks, having BABIP mean so much more than it should and guys like Griffey being almost immediately useless (even Murray was useless at the beginning of the game - he should have been good for at least a couple weeks).

I know none of us want Darren O'Day and peak David Robertson running out there and its ok for OOTP to have HOF starters in the pen but months old Kevin Brown should really not be better than Walter Johnson either.

The real HOF is flawed I think we can all agree, but if none of the best players in this game aren't named Hank, Willie, Ted, JR, Stan, Mickey or Babe doesn't that say that the calculations, algorithms or whatever need some tweaking? None need to be meta ("meta" should not even exist but whatever), there are still plenty to choose from to make teams have variety. Most of the above were / are usable but Billy Hamilton should definitely not be better and arguably neither should Gwynn. Rose should never have lasted so long either.

Anyway, there isn't a need for a couple players to be on every team especially when a lot of them do nothing more than hit singles. If power was more of a factor - which I think I remember them trying to do but not sure it worked - then there would be more teams with different players on them.

[chazzycat]

Originally Posted by frankbama

Was not aware there was a penalty for DH so learn something every day. Is there a reason for that? In "real life" as a basis some hit better as a DH without having to field and others have said they hate it since playing the field keeps them in the game and not thinking about hitting as much. What does the game infer about this and its effects?

Yep, it's because it exists in real life. It was originally analyzed & described in "the Book", which is required reading for anyone attempting to understand sabermetrics. I'm not sure if the game infers anything beyond the fact that the penalty exists.

[ncap99]

Originally Posted by frankbama

Anyway, there isn't a need for a couple players to be on every team especially when a lot of them do nothing more than hit singles. If power was more of a factor - which I think I remember them trying to do but not sure it worked - then there would be more teams with different players on them.

It was a point of emphasis this year but obviously not enough. It just seems to me that the drop off is too severe when scaling due to normalization. I think people look at a card with 100 contact and consider that a high amount - and when compared to all players, even non live perfects, it is. The issue is that if I used my roster as an example (I have zero chase cards, running FH3 with the standard fair Maddux, Rixey, etc) the average contact rating of my starters vR as a is 110.56. 100 contact rating is bad now.

The meta is high contact, and it just reinforces itself. In a balanced game you'd be able to say - hey, everyone is running a bunch of Tony Gwynn types so I'm going to run boppers out there, and because normalization my 150+ power guys are going to jack so many bombs it won't matter that they hit well below league average. In the current unbalanced way it works they hit so far below the league average that it more than counters the additional HR production. I'm sure someone with an advanced math degree could try and figure out the HRs required for a .200 hitter to make them outproduce the expected runs added compared to a guy hitting .300 with everything else (walks, Ks, XBHs, etc) being equal, but I can safely say it is higher than what you could realistically hope to get with any consistency.

So basically, the fall off is so steep it makes these power cards borderline unplayable. If those same ratings were getting you a .250 average then maybe it becomes viable?

I think the most simple way to say it is that if you took 100 power away from Babe Ruth and left EVERYTHING ELSE the same, he'd be a better card - how does that make sense?

[frankbama]

Originally Posted by chazzycat

Yep, it's because it exists in real life. It was originally analyzed & described in "the Book", which is required reading for anyone attempting to understand sabermetrics. I'm not sure if the game infers anything beyond the fact that the penalty exists.

True, though there aren't any injury or age progressions in this game and their defensive abilities are what they are so guys like Edgar, Papi or other full-time DHs shouldn't be penalized for playing their primary position just because technically they can stand near first base.

I'm nitpicking here I guess but it's in between coffee-time and happy hour so thought I'd comment.4

[chazzycat]

The penalty doesn't have anything to do with aging or defense. The research looks at the same players, hitting both as DH and non-DH during the same years.

Here is a slide from this year's SABR conference using real numbers from the 2020 season. The research has been re-examined several times since "the Book" came out and they keep finding the penalty every time.

[frankbama]

Originally Posted by chazzycat

The penalty doesn't have anything to do with aging or defense. The research looks at the same players, hitting both as DH and non-DH during the same years.

Here is a slide from this year's SABR conference using real numbers from the 2020 season. The research has been re-examined several times since "the Book" came out and they keep finding the penalty every time.

That's aggregate and it only shows players that played as a DH and a non-DH in the same season. Can't really trust 2020 data but I'm assuming findings would be similar for prior years as you say. That said, full-time DHs wouldn't play in the field enough to qualify in that situation. Ortiz, Edgar and Nellie would not hit better if they were playing the field every day. In our game most people have full-time DHs set even if platooned. I guess there's no way to separate that though from a rotation type in the game.

We also know that players DH when they can't play the field effectively due to injury which would bring down those averages if they're not 100%. Definitely read that in there somewhere too.

[Syd Thrift]

Originally Posted by Hertston

True, of course, but I tend to agree with the OP about Williams' performance comparative to Collins, Brett, Simmons, etc. All are facing the same pitchers. In purely historic terms I would expect only Ruth to have consistently better hitting results, although I've no idea if the meta card actually does/will.

Many congrats on your return to PL, BTW. Sorry, hadn't noticed that before.

I can't speak immediately for Choo-Choo Collins or Bucketfoot Al but Brett in particular struck out in 8% of his PAs (compared to Williams' 9%) in a much heavier strikeout environment. And now, because I have BBRef open and don't believe in edits (except to add this for clarification)... Collins played in a similar, although much more varied, K environment and struck out around half as often. Simmons I can't account for except that maybe the Simmons card is one of those super-duper-pooperstar cards they come out with where the player is maxed out beyond their regular ratings.

[chazzycat]

Originally Posted by frankbama

it only shows players that played as a DH and a non-DH in the same season.

Yeah, that's the idea...isolate the pertinent variables...if they did it any other way, it would just be showing the effect of aging, not DHing.

[Syd Thrift]

Originally Posted by ncap99

It was a point of emphasis this year but obviously not enough. It just seems to me that the drop off is too severe when scaling due to normalization. I think people look at a card with 100 contact and consider that a high amount - and when compared to all players, even non live perfects, it is. The issue is that if I used my roster as an example (I have zero chase cards, running FH3 with the standard fair Maddux, Rixey, etc) the average contact rating of my starters vR as a is 110.56. 100 contact rating is bad now.

The meta is high contact, and it just reinforces itself. In a balanced game you'd be able to say - hey, everyone is running a bunch of Tony Gwynn types so I'm going to run boppers out there, and because normalization my 150+ power guys are going to jack so many bombs it won't matter that they hit well below league average. In the current unbalanced way it works they hit so far below the league average that it more than counters the additional HR production. I'm sure someone with an advanced math degree could try and figure out the HRs required for a .200 hitter to make them outproduce the expected runs added compared to a guy hitting .300 with everything else (walks, Ks, XBHs, etc) being equal, but I can safely say it is higher than what you could realistically hope to get with any consistency.

So basically, the fall off is so steep it makes these power cards borderline unplayable. If those same ratings were getting you a .250 average then maybe it becomes viable?

I think the most simple way to say it is that if you took 100 power away from Babe Ruth and left EVERYTHING ELSE the same, he'd be a better card - how does that make sense?

SORT of. The meta is high BABIP because that's the one major number where you'll find a lot of variance in these top-rated star players. Pretty much everyone has at least average power, lots and lots of great hitters walk a lot, very few playable guys even have bad defense at this level... but while everyone, too, hits for a high average there are many different ways of getting there, and in particular the "hits lots of HRs and doesn't strike out a ton compared to their league" is actually kind of well-worn.

On the other hand, BABIP is a hidden rating. Contact is an amalgamation of Avoid Ks, which of course you do see, and BABIP, which is hidden. That said, different players have it in varying degrees and BABIP isn't something that tends to be limited by pitchers' Stuff the way K rates are, so guys like Ichiro still excel. IRL if you literally played with a land full of nothing but superstars, the Ichiros and Rod Carews of the world would probably be nerfed pretty heavily by rangey fielders, but the way OOTP adjusts statistical outcomes means that even in a league packed with 80/80 fielders up the middle, all those singles have to come from *somewhere*.

And with BABIP, there really isn't the same kind of baseline of "everyone you might otherwise play is at least average". Like, Reggie Jackson played in the same general BABIP environment that Rod Carew did (the league averaged .281 for both over their careers) but Carew BABIPed almost 70 points higher. Similarly, Ichiro BABIPed almost 40 points higher over the course of his career than David Ortiz did. There's still room for big, big differences, much more, I think, than any differences between the best of the best in terms of K rates and even HR rates. So it becomes a combination of these two effects: batters who are otherwise really close to equal in terms of avoiding Ks, and whose differences in HR rates are only worth, say, an extra 25 to 30 hits a year, get an even bigger boost from the game engine, and there isn't a way to negate the BABIP boost the same way that you can soften the effects of high HR hitters by having a squad full of high Movement guys or walks with high Control (and for a guy like Ichiro who never HRs and rarely walks, what's a high Control and Movement guy going to do to him, not walk him or give up a HR twice?).

I guess there's a good question here as to whether the game engine should just run the same for superstar leagues as it does for normal, everyday ones. My first inkling is to say "yes" because, well, doing other crap would either cause a lot of work or would make the stats being produced so far off the "norm" that they might not be very useful to look at (imagine a league that averages 16 Ks per game due to everyone having like Sandy Koufax as their #1 starter, the league BABIP being like .240 because of all the great fielders... and a league ERA of like 2.00 where the scoring is all with walks and homers... setting aside the fact that you couldn't really do much with the stats, is that even a fun environment?). That said, the game makes some big compromises to keep the levels balanced, and BABIP is one of the places where this shows through.

[ncap99]

Originally Posted by Syd Thrift

On the other hand, BABIP is a hidden rating. Contact is an amalgamation of Avoid Ks, which of course you do see, and BABIP, which is hidden. That said, different players have it in varying degrees and BABIP isn't something that tends to be limited by pitchers' Stuff the way K rates are, so guys like Ichiro still excel. IRL if you literally played with a land full of nothing but superstars, the Ichiros and Rod Carews of the world would probably be nerfed pretty heavily by rangey fielders, but the way OOTP adjusts statistical outcomes means that even in a league packed with 80/80 fielders up the middle, all those singles have to come from *somewhere*.

A very good post, I especially like your analysis of why BABIP is meta - it applies regardless of the pitching or defense, while other offensive aspects can be schemed around.

Based on other things I've read (including in this thread), contact is a combination of AvoidK, Power and BABIP not just BABIP and AvoidK? I think that is the key point with cards like Babe Ruth - raising power without raising contact lowers BABIP which in turn makes the card unplayable because the curve the league grades against is so tight. It is like getting an 87% on a test and having that be an F because everyone scored between 87-100%. The curve needs smoothing or they need to massively inflate contact ratings of power hitters which would make the cards just look silly and unintuitive. Or just separate power from the contact formula entirely.

I guess a lot of my points are based on the assumption that power is used in the formula for contact too, and if that isn't the case then it really becomes difficult to explain the relative performance of Babe and Ted.

[Fabtron7]

You're correct... the contact rating is an amalgamation of BABIP, Avoid K, and Power.

[AdmiralTrey]

Originally Posted by chazzycat

It's much smaller, more like 3%.


Yep the game has always been harsh on HR-hitting sluggers who also had great contact skills & batting averages. Those types of players are just so elite & so rare (Ruth, Williams, Bonds) that I don't think the algorithm for translating stats to ratings handles them very well. Across baseball history, sluggers tend to strike out too much and have lower averages, which I believe is why the system is set up that way. It handles the majority of players pretty well. But in these extreme cases it just falls apart, and of course these "extreme" cases are the players we love the most.

This year actually seems a bit better than previous, for what it's worth. Ted and Ruth are both getting significant play in the highest levels despite these perceived shortcomings.

The DH penalty is only 3%?? That's good to know! That's not nearly as drastic as I've been thinking it was. Thanks for that info.

My Ted Williams is still putting up fine numbers, so I agree that this year is working out better than last year. Still waiting on a 100 Bonds card though...

[dbqs]

My impression is the heart of the problem with power hitters in PT is less the actual weighting of the ratings of the cards and more the normalization of the league environment. It could be more interesting to see a little more looseness in the possible outcomes depending on the distribution of the players in the league, i.e. if every team were running Ted and Babe we'd see a ton more HR's rather than everyone's Ted and Babe hitting fewer HR's to meet the baseline.

[Syd Thrift]

Originally Posted by ncap99

A very good post, I especially like your analysis of why BABIP is meta - it applies regardless of the pitching or defense, while other offensive aspects can be schemed around.

Based on other things I've read (including in this thread), contact is a combination of AvoidK, Power and BABIP not just BABIP and AvoidK? I think that is the key point with cards like Babe Ruth - raising power without raising contact lowers BABIP which in turn makes the card unplayable because the curve the league grades against is so tight. It is like getting an 87% on a test and having that be an F because everyone scored between 87-100%. The curve needs smoothing or they need to massively inflate contact ratings of power hitters which would make the cards just look silly and unintuitive. Or just separate power from the contact formula entirely.

I guess a lot of my points are based on the assumption that power is used in the formula for contact too, and if that isn't the case then it really becomes difficult to explain the relative performance of Babe and Ted.

Yeah, Power comes into play as well but only to the extent that HRs increase a player's batting average. So, a guy with 40 HR power will see his CON rating raised the equivalent of 40 hits compared to a guy with 0 HR power.

[ncap99]

Originally Posted by Syd Thrift

Yeah, Power comes into play as well but only to the extent that HRs increase a player's batting average. So, a guy with 40 HR power will see his CON rating raised the equivalent of 40 hits compared to a guy with 0 HR power.

Do you think the issue is that power just doesn't generate enough HRs, which would then lift BA without effecting BABIP? Just from clicking around the 5 people running Ruth in PeL are on pace as a group to average about a .250/26 HR season. Doesn't really seem to be that different than Aaron or Mays who are 65+ points lower in power. Ruth has 25 points more power than any player in the game, if he is the A+ grade on the curve he should produce A+ HRs in the 40+ range consistently I think. 15 HRs more HRs in 500 ABs is worth 30 points of average. It just seems the way the normalization works and the way ratings are considered squishes the range of 80 to 220 power down to about 20 total HRs.

Or maybe it's like some people have said in this thread, everyone throws as much movement as possible and a lot of people (myself included) suppress HRs with park factors - at some point it just isn't worth trying to run uphill and you just say screw it and run gap hitters out there.

[professor ape]

Originally Posted by dbqs

My impression is the heart of the problem with power hitters in PT is less the actual weighting of the ratings of the cards and more the normalization of the league environment. It could be more interesting to see a little more looseness in the possible outcomes depending on the distribution of the players in the league, i.e. if every team were running Ted and Babe we'd see a ton more HR's rather than everyone's Ted and Babe hitting fewer HR's to meet the baseline.

Agreed

[Tinkerman]

A nice little side effect of the new cards is my Boggs playing very well over the last few weeks. Never gonna match the overall contribution of Brett, but I expected him to be defunct by now!

[OMGPuppies]

It’s sad that based on the game engine that certain players can never really excel at the higher levels in PT.