This really isn't about Addie Joss but we'll start with him because he's the best value in open leagues.  The reason he and so many deadballers are so effective in whatif is because of the way whatif normalizes between eras.  Honestly if Addie Joss ever pitched to Albert Pujols, you know the ball would end up in the bleachers.

I don't want to get too mathematical but the way this hypothetical matchup works in WIS is to take 


To calculate any particular event  in any pitcher/hitter matchup (what many people mean by "normalization"), WIS performs a calculation that includes:
-the hitter's actual (event);
-the hitter's league average (event);
-the pitcher's actual (event) against; and
-the pitcher's league (event) against average

Although I don't have the exact formula any more (it was once on the forum somewhere), I do know that home runs occurred about once every 10 games back in Joss's day, compared to one per game in today's game.  So start by dividing Pujols HR probability by 10.  But since it is a cross between Joss' year and Pujols' year, the net effect is divide by 5.  So a full season of Pujols vs Joss would result in Pujols having less than 10 HRs per year.  (By the way you can test this in sim-matchup)  That is the major fault with WIS.  That is why WIS has brought upon the birth of dead ball.

I have the fix, which I offered to WIS some time ago, but maybe if one of you more well-known and better tolerated users would latch on and convince them.  The answer is to replace the current 4 factors in any event with these:

- the hitter's standard deviation vs league average (event)
- the pitcher's standard deviation vs league average (event)

In this way, a guy who was among the best in his league in preventing home runs in 2011 (like Matt Cain) would be as good as dead-ballers in preventing home runs.  More importantly, dead-ball pitchers would no longer dominate WIS.

Who's on board?

For those who want to understand about standard deviation, I have provided the link below:

http://en.wikipedia.org/wiki/Standard_deviation

Interesting and thoughtful, thank you.  How would this take into account the "regression towards the mean" baseball has experienced over time?

SpotSell, I'm not exactly sure what "regression towards the mean" is?  Can you offer more details?  

The formula I propose has only two points in time, the pitcher's year and the hitter's year.

Basically it means the deviations between the the very best and the very worst vs. the league average has greatly declined over the decades.

Here is a piece on it, I could not quickly find the original 1996 piece by Stephen Gould.

http://www.thefreelibrary.com/Maybe+There+Were+Giants%2c+Or+at+Least+Outliers%3a+On+the+.400+Batting...-a058517919

I hope that you've also included that in the dead ball era, what is now a ground rule double, was then a home run. After all, if we are truly attempting to find a fair comparison, this cannot be left out. 

Additionally, how many dead ball era players' hard hit ground balls were fielded cleanly as opposed to those hit by the "rock" they use today??? Besides the speed of the ball coming at an infielder, said speed would also get through the infield holes much more often today. (Especially on artificial turf, which didn't exist in the dead ball era and should be accounted for if we are going to make it all truly fair.) NON dead ball era batting averages should be adjusted accordingly as well if we want all to be even. Then again, exactly HOW many dead ball era singles would be fly ball outs today and how many dead ball era fly balls would be today's home runs??? (Can't wait to see how THAT is calculated.)

All I'm trying to say is that if all should be made even and fair with today's pitchers as compared to those of the dead ball era, then the same ought to be done with hitters. Otherwise, things would be uneven. After all, dead ball era hitters don't HIT home runs like today's hitters do in this game, and dead ball era pitchers don't give up home runs like today's pitchers do. Today's batters get their hits in a different way than the batters of the dead ball era did. AND today's pitcher's get their outs in a different way than pitchers used to.


While I appreciate your passion and that of others to make things even...In my humble opinion, it can't be done!!! No matter what, there will always be some dispute...

Just sayin'

I agree with you harrelson, I'm sure that  the integrity of realism as it pertains to what actually happened is something that WIS would like to maintain but the overriding concern should be to reduce the marginalization of a full era of players at the expense of another. Especially since it stands to reason that most people can relate better to modern day events.  Your suggestion is a good one.  Deadball pitchers are moving 2 full standard deviations from the mean when the pool is expanded to all players. And as SpotSell said there is a regression toward the mean. In almost every aspect of achievement in life that usually says that the talent pool has actually improved over time (more people approaching their optimum skill level, because perfection is finite.), further making the argument for modern day players. 

The most enjoyable leagues I've been in recently have been themes that exclude deadballers--especially progressives. It's nice to be able to draft a HR hitter knowing that he has a chance to hit somewhere close(r) to his real life total. I drafted the '83 Ron Kittle in a progressive--his only skill is being able to hit HR. A guy like that is useless against deadballers. But in this progressive, he's an integral part of my team. I could be mistaken, but I'm pretty sure his APBA card from that year has double 1's (totally unrelaed, but felt the need to reminisce back to the low tech days of "simming"...)

This would be great for HRs, but it would really create some skewed results. To expand on what Spotsell is saying about regression to the mean... here are a couple of fictional examples that are loosely based on reality:

-Babe Ruth was something like 10 standard deviations above the mean for HRs in 1919. Probably 1920-21 too. Doing this would make him good for approximately 100 home runs every single season. 

-No MLB hitter was more than 3 standard deviations above the mean for HRs in 2008-2009. In a world like this none of them could ever hit 30 HRs in a season.

-Mariano Rivera was about 20 standard deviations below the mean for HR/9 in 1996. He would basically be throwing a brick up there.

The problem that you're talking about is the fact that the WIS normalization method works pretty well for everything EXCEPT home runs. 

I don't know what the answer is. Just use HR/9+ and ignore the raw numbers? Then if you played a single season 1910 league, some random guy would hit 40 HRs and I don't think that's what we want. 

It seems to be a bigger problem than I imagined.  Perhaps something could be done like golf handicapping, the season strength would equal the course rating?

Well, they have normalized fielding now for every player based on the hitter/pitcher dynamic. Why can't they do something similar for HRs? It would be as easy as taking the HR rate for the hitters season and the HR rate for the pitchers season, averaging these together as a baseline for the HR rate of the AB, and then using the players real life numbers off that baseline.

Using completely made up numbers:

You have a hitter from a season where the HR rate is 5 HRs/100 AB (modern era) and the pitchers rate is 1 HR/100 AB (deadball era). This would make the current HR era for the AB play like 3 HR/100AB. Then, you would use the actual normalized stats of the hitter/pitcher involved from there like they do in the tree now.

It may still skew HR downward, but would create a normalized reality like they do with fielding currently.

Just thinking out loud here, maybe if WIS did two rolls.  The first would be a roll for a HR, using the STD DEV method above, if HR boom goodbye, if not, then the 2nd roll:  This time use the current method and decipher result of play from there.  This way the change would only affect HRs and the regression problem would be limited to HR hitters.  I think you have to accept some inequalities when playing two different eras against each other.

I am very new here, so please take my comments as presented with humility, not my coming in like a bull in a china shop.
First, let me say that I really appreciate the high intellectual level of the discussions taking place on this and some other forums here, including some about the site itself.
Naturally as a first season "rookie" I am happy to find it and it will take a while for that to wear off, though having read through a lot posts the past two days to get up to speed, I see that there is some legitimate disgruntlement among players here about how results work and whether they are reasonable or not (ignoring the questions being batted around about the relation of the site's management to users which is a different question I don't feel qualified to address yet). 

One issue coming up is about the deadball era pitchers' dominance or the Joss problem (by the way my new teams beat him twice in the past few days in different leagues). I admit that the one concern I have is that  this will not be a lot of fun if we can game the game by simply putting in a bunch of deadball era pitchers with 400 innings pitched and ERAs, that, despite some of the legitimate points made in the counter-arguments about fielders' gloves, ground rule doubles and so forth, clearly are lower than they would have been in the 1930s or the steroids era.  

I wanted to bring up a few issues that I think address the three most important aspects of the problem as I have come to see it in the past few days (again, admitting that this is new to me, but also evaluating how much time I want to commit to something if it won't be as enjoyable once I get to know it better as it has been so far). 

1) It is important the we enjoy this. I think that there are two aspects to that: winning, and enjoying playing this. We can't pretend it is not more fun to win than lose, but it is also true that after a while winning because you figured out that a certain combination of strategies can't be defeated must get boring. So my first suggestion would be to try to campaign to get users to treat this more like baseball and less like an online game. That means that in reality you would not have two copies of Addie Joss and two of Mariano Rivera and then just use them over and over to wear down opponents. It is true that sometimes in real baseball managers with great confidence in a few pitchers just went with them a lot (2001 Arizona Diamondbacks being a recent example) but even they had to turn to relievers and a third starter every third day. So if we just decide that the fun is this is winning at BASEBALL, not at overcoming a math problem, we can tone down the overuse of a few starters and make sure this stays competitive, or does so without everyone using the same players and strategies. 

Frankly, with the exception of a few players that I am convinced would have been great anyway - Mathewson, Walter Johnson, Pete Alexander, Ty Cobb, Honus Wagner, Hornsby, maybe a couple of others, I don't trust the deadball era players. The game was too different. Plus, I have no emotional stake in them, the way I do the players i named above, plus the Babe Ruth era and on, partly because the teams aren't even around anymore so who cares how great they were - no one today glories in the accomplishments of teams that no alive ever saw play even in their contemporary versions. That is not true of the Detroit Tigers, Pittsburgh Pirates, or even teams that have moved around like the Dodgers, or changed names like the Texas Rangers, whose fans can include some kid who knows that Walter Johnson and Nolan Ryan once pitched for the team he or she roots for today. 

2) Travel. This is a missing element even in the brilliant discussions, on both sides, about the standard deviation etc. Travel may have been physically harder in the deadball era - buses and trains - but it did not include many time zone changes. This is hard to account for, but no one had to go west of St. Louis to play back then. 

3) The history of baseball cannot be easily divided into two camps, deadball and post-deadball because themselves were not homogeneous eras. Leagues rose and fell, gloves were invented and improved, the curve ball invented, and so forth even in that era. Since then we have the post deadball inflation of scoring in the 20s and 30s, We then have a leveling off right before the war (in which DiMaggio and Ted Williams and Hank Greenberg among a few others stand out exactly because of this - ask yourself how many players from the 1930s still resonate today emotionally, yet those from the immediate prewar era still do). Then the years 1941-45 for which we can't trust any statistics. Then what we might think of as a reasonably "normal"era from 1947-say 61 except that of course a change arguably much greater than the end of the deadball era occurred whose statistical importance is extremely difficult, or so easy as to be disturbing, to calculate. Then the mound is lowered and there is expansion. Then 1968 you get absurdly low ERAs and batting averages, only to have them go back up a year later when the mound is raised again. Then the 1970s when I was really becoming a serious fan and what seemed then like normal now seem like low batting stats and scores - Graig Nettles led the league with 32 homers once. Then comes the home run boom and steroids, now who knows...

So it is not so simple to just divide the world into pre-and post Ruth without admitting that every period of baseball player has their statistical anomalies. Which brings me to the big one:

4) The Caucasian Leagues. Or the Majors as they were then known. Here is it in a nutshell, the deadball-post-deadball comparison probably pales compared to another, which is either incalculable or very easily calculable - a certain percentage of all players in the majors before 1947 - let's say somewhere between 10 and 20% - I would settle on just a little below 15% to be nice about it - were bums who didn't deserve to be there on the merits, and only had jobs with major league teams because 15% of the population was not allowed to play.

Or to be more precise, were not allowed to compete with them. They played anyway and from all accounts a high quality game, one where Jackie Robinson learned to play the way he did when he got to the majors and where Hank Aaron learned to hit. Not every player in the Negro Leagues was that good either of course, but we can figure that by definition every statistic for the majors before 1947 for pitchers and batters is inflated by - let's say 10% to be on the conservative side. That means that 10% of the pitchers that both Ty Cobb and Babe Ruth faced were minor leaguers. Not only, but the 10% of black players missing were by the definition the best of that 15% of the population. So instead of .367 lifetime, Cobb hit .330. Very, very good. Ruth hit 643 homers, or 20 less than Willie Mays (though in the argument about the greatest player remember that Mays never pitched a shutout in the Series).  Joss' and the rest also likewise have ERAs lower by 10% than they would have been. 

Now, one way to compensate for this would be to do what I just did, in a rough and undeniably over-simplified way to just decide how much we should cut the pre-47 stats. By the way this whole thing puts the issue of Barry Bonds, Mark McGwire etc. stats in perspective doesn't it? I don't like them holding the record personally because of the steroids thing, but it is not the first time someone used a completely unfair advantage to set records. 

Another way is perhaps more interesting: include the Negro records and stats, inasmuch as we have them, and many we do have, as equally reliable as those of the Caucasian Leagues. Addie Joss or Satchel Paige?  Now it gets interesting, and some of the advantage of all pre-1947 players, not just the deadball era ones, is compensated for by their own contemporaries. 

Anyway, as anyone who looks my teams will see, I have on the one hand learned already much from my competitors (you all) whose intelligence I have already come to respect greatly. My initial attempt to deal with the whole innings pitched requirement to compete through using 5-man rotations ended in fiasco in one week, as pitchers with 200 innings or a little less quickly showed wear and tear - though some of this I have learned is due to this weird timing of fatigue hitting after a long winter's rest after one or two starts - Jim Palmer pitched opening day, went 6 innings, this from a season where he had over 250 inning pitched, and he is already listed as worn out. But this aside (I digress) I saw that 5-man rotations, which in real baseball I have always hated as unnatural, and Yankees fan though I am and worshipper of Mariano and the gods that sent him to the Bronx, I hate closers as an institution. Not the individuals obviously.

So in keeping with my own first suggestion, I am going to go with my own strategy: four man rotations with good amounts of innings but nothing super-human; my own assessment of how much I trust the greatness, or quality, stats aside,  of those before my time (Ted Williams, Dimaggio, Ruth, Paige, Josh Gibson, Mathewson, Walter Johnson - yes, Jack Chesbro, who knows? Maybe, some of the others, not so much - arbitrary, but my arbitrary);: I will load up on good quality long relievers and try to use good relief pitchers like Lyle, McGraw, Gossage, Clay Carrol, and Bill Lee among others to come in when the starter can't finish the job - as they used to: sometimes to get the last two outs in the 9th with people in scoring position in a close game, sometimes in the 6th inning and pitching the rest of the game. I know what kind of hitters I like - power, average, don't strike out, good OBP. Take a look at some of the power hitters from the 40s and 50 and their SO ratios, they are more like contact hitters' today, then look at Reggie Jackson, Richie Sexson, etc. some of these guys strike out once a week. Don't like it. 

Baseball is like chess often, that is part of what we love about it. Chess is fun too because it is a challenge, a puzzle, and is more likely to be solved by someone that has a mathematical mind than baseball, though since there is another player even chess cannot be guaranteed. But they came up with a computer that beat Kasparov more times than he beat it. OK. But I wouldn't go to see it hit a baseball, or watch robot baseball sometime in the future, with 900 foot homeruns. What would be the point. So let's keep this to baseball with its element of chess and we can probably enjoy it more. Want to use Joss because you like him? Get him and use him. Once around the rotation. If you spend enough on him then your numbers two and three will be more like, say Don Drysdale and Bert Blyleven or Frank Viola, or David Cone. Or fill in the blank yourself. Good players. I like to play mostly the ones I have seen play and then learn about ones that have become good more recently, and add one or two from long ago that I have a gut instinct can play - Yogi and Jackie and Stan Musial. 

Anyway, I repeat that while this may seem like I have barged in here after a very, very short time playing, I do so out of respect for the seriousness of mind of my fellow competitors. I hope to see you on the field. play ball, 

Good rant...and I can tell you are passionate about the game and history of it.  Please check out theme leagues and progressive leagues. I think you will find them very much to your liking. Theme leagues have specific rules that (given the right league) will completely solve the problems you have with deadballers (or any other era) and progressive leagues are even more fantastic about having all the players be from the same season.

Just because open leagues are the default, does not mean the game only revolves around them. 

To be honest, I've played on this site since 2003 and have yet to EVER play an open leauge.

Thanks for the compliment last week, but you must understand that unlike chess, people are here for varying reasons. Here, there is more tolerance. As long as it is in the parameters of the rules, it is fair game, and people feel free and are welcome to do it. I notice that you took offense to someone using 2 Joss's, 2 Bip's  or 2 Rivera's in a common league we're in.  If that's what they enjoy doing, so be it.  You can't control them, there is no ethics control board.  You mentioned chess, chess is less forgiving. If someone even makes a move that a player feels represents that his rating class is even three levels below him, he like most strong chess players will seek never play this individual again because there is nothing to be gained by being in his company and someone who is losing is expected to resign or offense is taken. There's nothing to be lost by reading a boxscore of someone that did something unconventional or novice like.  Whereas in chess the pain is that you will never get that time back.  Just think of the difference between Fischer in his Championship match against Karpov, when everything from the sound to the lighting to him feeling the Soviets were feeding him subliminal sugestions almost stopped him from playing and how everything needed to be uniform and precise. It's more relaxed here. In chess a thousand people have analysed every variation of the  Sicillian defense sometimes 17 moves deep or more, here the waters are unchartered, there is no "Bible" and you will see many strategies of trial and error to collectively better the site. GL

Thanks Frazzman and Crimson both. Frazzman, I will check out the other kinds of leagues, but first need to get more experience playing. I have a bunch of teams playing now in different leagues and want to play out a full season with them to see how variations on my strategic approach - including my preference now to eschew closers altogether, and go with 4-man rotations, and six long reliever-closers by committee to see how it might counter the currently dominant 3-man 300 IPs-plus approach. If it fails, I and we will know. Which is why Crimson, I appreciate that you have a point, but maybe my comments seemed more critical of actual individuals and how they play here than I meant it. I was not criticizing people, or suggesting that everyone doesn't have every right to the strategy they want to pursue, merely that on an adjacent forum here people are saying they are getting bored with the same strategies, which so many teams seem to use that mine are scheduled against - 3 man rotation, mostly deadball pitchers, and Vince Coleman. 

Of course people can play as they want. But if a newcomer enthusiastic about the site finds those playing a while complaining to each other that they are bored, it makes an impression. My goal will be to create competitive teams. If it doesn't work, we may find that I am loading up with Joss and co. as well in a year out of frustration at not winning. But I at least know that I am rooting for my teams, and when Eckersley pitches well, or Spaceman Bill Lee gets a save to beat Koufax, I am having a blast. I just hope everyone else is too, and that I have not left any hard feelings, as that was far from my intention.