ATG9 Salaries - the Egregiously Overpriced Cards

Thanks guys for chiming in.

I agree with Hack Wilson that the Babe's 16.47M is perhaps best used higher in the lineup than BAbe's 13.07M or 13.03M cards (although if I were to use him, I would probably build my lineup to set him as #2 rather than #1.

Hack also alludes to the Babe's speed with the 16.47M card, and I think it's part of the reason his card is snubbed relative to the 13.03M: the peripherals of the Babe's 16.47M card are better to the point of making a substantial difference.

Here is the rating of the Babe's card with games played in US Cellular type of park (low ballpark singles, high ballpark homeruns).

First, looking only at the offensive card, not considering running, defense, or injury-risk...the pure offensive card reading (WAR and salaries based roughly on 150 games, 648 chances precisely):

Card...........WAR........estimated salary
Babe 16.47M...11.7 WAR...15.73M
Babe 15.10M...11.2 WAR...15.20M
Babe 13.07M...8.3 WAR....12.34M
Babe 13.03M...10.1 WAR...14.15M

So the 13.03M offensive card is not far from the 15.10M and looks like a good buy.

Now, this is how the WAR is affected after considering the running game, the defensive side, and factoring in injuries. Remember that only Babe's two costliest cards are injury-free---the two 13M cards are expected to miss 6 games (and a few more at-bats in games where they are injured). I consider for the ratings a bench player worth 0.7M worth his card...for outfielders, that would be cards like Tom Paciorek and Frank Delahanty. In order words, the ratings below consider 162 games from Babe's 16.47M and 156 games from Babe's 13.03M supplemented with 6 games from Paciorek:

Card.......Off WAR.....run-adjusted...def-adjusted...inj-adj......estimated salary...value
Babe 16.47M...11.7 WAR...12.6 WAR...12.5 WAR...13.3 WAR...17.29M........+0.82M
Babe 15.10M...11.2 WAR...11.9 WAR...11.0 WAR...11.7 WAR...15.74M........+0.64M
Babe 13.07M...8.3 WAR....8.6 WAR......8.6 WAR....8.7 WAR....12.70M........-0.37M
Babe 13.03M...10.1 WAR...10.4 WAR...9.4 WAR.....9.5 WAR....13.54M........+0.51M

Babe's 16.47M card gets a slightly bigger run-adjusted bump (+0.9 WAR) over Babe's 15.10M (+0.7 WAR) despite both cards having the exact same running ratings because he's more like to reach first base and have the running game get going compared to the Babe's 15.10M.

That said, the +0.3M difference in favor the Babe's best card compared to his 13.03M card is rather small, so I would consider other issues before choosing one or the other. The Babe's 16.47M best lineup position, in my opinion, is the #2, so he needs real solid #3 and #4 threats. Babe's 13.03M is more a typical #3 or #4 hitter, so perhaps the better choice is the lineup is thin offensively. Also, Babe's 13.03M has a better arm, so looks more like a right-fielder compared to the 16.47M whose -1 arm makes him better suited for the left field. And as I wrote, the 13.03M has an injury risk, so the quality of the bench must also be considered before going with one or the other.

Very interesting, and yes, the non-injury factor for the 16.47M card is a major plus in my view. I had the Babe batting second in the beginning of the season (after Buddy Myer), but then moved Babe up to first with the 11.83 Stan Musial card hitting behind him and Justin Wilson after that. My team started winning, so I kept the Babe there. The other thing, it's a DH league, so the ninth hitter is a professional hitter, not a pitcher, so more opportunities for Babe to drive runs in from the one hole. In fact, I try to put a slightly high OBP, cheap guy or platoon (I love platoons from the fifth or sixth spot down) in the ninth spot for this reason.

Lineups depend on the players you have, ballpark (I'm in Yankee '71, 12 for lefty homers, so not quite a bomber park), but I do think putting your best three at the top is always good, and that can include an awesome platoon, like Terry Harper and Berbie Carbo, or Lee Mazzilli and Ken Smith, in a particular park, if you want to maximize hitting or on base that way.

Anyhow, back to the original thrust of this post. Like others have pointed, I'd say out that ratings/value are fluid objects depending on the context. So many variables exist -- opposition, ballparks, lefty or righty, rosters, match-ups (!) -- that a player's value changes given the circumstance, league, team, park, pitcher. The seemingly fixed way to understand the player value is through our myriad rating systems, which seem to disagree quite a bit, so that may leave questions unanswered -- though I find ratings like Joe the Jet illuminating and compelling enough to learn something I did not know before, if not perfect, and I don't expect much to be perfect. Ultimately ratings reflect how much weight someone places on some particular trait of a card. It's arguably mysterious because people seem to have different interpretations about how much weight should be given to this or that trait or environmental variable.

So many ratings exist, so ask yourself what this means.

So many ratings exist, so ask yourself what this means.

What this means is, all the confusing detail aside, the salary structure is fundamentally broken and that problem has been the leading driver in the decline of participation in the online game.

If player value is so fluid depending on the context, how can we ever have correct or even a largely accurate salary structure? I guess it depends on what we mean by fluid and the extent of it.

Very interesting, and yes, the non-injury factor for the 16.47M card is a major plus in my view. I had the Babe batting second in the beginning of the season (after Buddy Myer), but then moved Babe up to first with the 11.83 Stan Musial card hitting behind him and Justin Wilson after that. My team started winning, so I kept the Babe there. The other thing, it's a DH league, so the ninth hitter is a professional hitter, not a pitcher, so more opportunities for Babe to drive runs in from the one hole. In fact, I try to put a slightly high OBP, cheap guy or platoon (I love platoons from the fifth or sixth spot down) in the ninth spot for this reason.

Lineups depend on the players you have, ballpark (I'm in Yankee '71, 12 for lefty homers, so not quite a bomber park), but I do think putting your best three at the top is always good, and that can include an awesome platoon, like Terry Harper and Berbie Carbo, or Lee Mazzilli and Ken Smith, in a particular park, if you want to maximize hitting or on base that way.

Who does Babe Ruth think he is? Anthony Rizzo?

I can't speak to a lot of the issues raised in this discussion, but I also hold to Marc's belief that positional adjustments while somewhat arbitrary, don't make an entire position over/under priced they are just applied equally to an entire position designed to balance certain defensive positions that get more opportunities and have the opportunity to produce more defensive value.

The point of a good pricing model is to pick out irregularities in the ATG prices. If my model values OBP more than the ATG pricing model, then those players as a group should be "underpriced" based on my evaluation, but the key is they will be under priced consistently as a group. Thu I can live with issues like the entire 3B group is over priced as long as the pricing is consistent. Where things break down is when individual hitters do not follow this model. I believe those exist in ATG and I think these threads on both over/under priced have flagged a lot of the most egregious ones.

In general I believe the 200X games are priced fairly competitively to deal with park fluctuations. However, in 200X you also are dealing with scarcity when pricing which adds a different dynamic and usually limits the ability to max/min your roster to your park. No such limitations exist in ATG when building from the size of the ATG player pool.

ATG9 needs to be repriced using a consistent pricing model. It doesn't have to satisfy everyone's individual models, if it did what would be the point of all the work for those of us who try to find values in the system to win. The prices just need to be consistent and I believe as Marc has suggest above, the pricing at a minimum could build in a premium for values based on certain extreme park adjustments. That's as far as I would suggest it needs to go. The better managers will always find the efficiencies in the model, they should just be a bit harder to find than they are now.

Where is the data that participation in the online game has declined?
Be very curious to see those numbers.

Certainly, the advent of introducing the rookie leagues was a solid move as many rookie owners got disenheartened by experienced owners dipping down and cherry picking the public leagues.

If participation is down, its not just the salary structure that affects it.
for sure the bias in STRAT against certain teams that I guess HAL doesn't like, is evident

For example---I know Baltimore Oriole fans who have stopped playing because of the HAL bias against that team.....the cards for Palmer etc are just awful. And I am sure some Phillies fans and Tiger fans would starting hollering as well.

And there are many cards that are simply not used because they are priced too high, or they don't reflect people's perception of what their value should be.

Here are some suggestions on improving participation:

1) Either reduce the salary of cards that are not used or eliminate them from the game

2) Solve the imbalance of NY players both in terms of quantity and quality---there are teams/cities that are basically ignored and/or minimalized.

3) Use a rating system to determine salaries that is consistent with stats/WAR/Usage---and avoids arbitrary determinations and biases

4) The injury factor for the Negro League players is simply false--only because effective stats were not kept back in the day. Most of those players, playing league games, barn storming games, and games in other countries, far exceeded the plate appearances of their major league counterparts. Fix this, its absurd.

5) Eliminate the 300 PA requirement for hitters--add players that the owner base would really enjoy having available, which would increase participation in the game--Dusty Rhodes and Hurricane Hazle are my personal quests...

If player value is so fluid depending on the context, how can we ever have correct or even a largely accurate salary structure? I guess it depends on what we mean by fluid and the extent of it.

Agreed. I don't think there is a holy grail for pricing. Every pricing (or equivalently, ratings) model has to make assumptions for certain variables or else there is never convergence to a deterministic solution. The biggest one may be salary cap. At high enough caps, no pricing matters. So as cap gets smaller, pricing becomes continually more important. You probably could make the case that pricing importance increases exponentially as salary cap approaches zero. And pricing importance approaches zero as salary cap approaches infinity.

Sure we can make relatively common assumptions writ large, like L-R splits, but it is just an average and an assumption. Matchups can be exploited at the tails of the distribution of event matchups.

That said, there are definitely some players who are either higher or lower than they should be priced, as gauged by how they are used (or not used at all) for the prevailing conditions in SOMO (caps, L-R etc). So I guess it depends what you mean by egregious...

I know Baltimore Oriole fans who have stopped playing because of the HAL bias against that team.....the cards for Palmer etc are just awful.

Sure, Palmer has two 6M-ish cards despite having twice a WHIP at or under 1.05, but did you consider that Palmer pitched in front of perennial def-1 such as Brooke Robinson, Mark Belanger and Bobby Grich? And Blair got 3 golden gloves at center field during these years too.

Don't forget that SOM builds their cards to reproduce team performance, not individual performance.