ATG9 Salaries - the Egregiously Overpriced Cards

My ratings are very different from Riggio's too. I see five boys on Riggio's egregiously overpriced list that my ratings have as steals already.

Well, I should put the emphasis that my ratings and Riggo's ratings are in agreement for 50 out of the 70 or so players he listed. So I would say that he's right on most cases.

For the dozen of players below or so, our ratings disagree.

Bargain value, not at all overpriced
4.73M Kingdom when used as dh (but even when playing the field, he would be just okay, not egregiously overpriced)


Priced just okay (value falls within 0.5M of the price tag)
6.92M Hill (because of a premium I give to cf arms -4 or better)
5.95M Sierra (because of a premium I give to switch hitters)
5.80M Tony Oliva (after correcting for Diamond Dope mistakes)
3.77M Don Buford (because of a premium I give to switch hitters)

Abreu appears slightly overpriced, but I would put him in the "just okay" category when used as lead-off (because walks are more valuable to lead-off, and Abreu has a lot of walk)

Priced just okay when used as dh (overpriced when used as fielder)
-10.64M Ralph Kiner
-6.59M Edgar Martinez
-6.01M Billy Hamilton
-5.78M Rusty Staub
-4.48M Fothergill

Priced just okay when used primarily at a secondary position
6.80M Pennington, when used as 3b
6.00M Stargell, when used as 1b---the 2.39M overestimation by Riggo must be a typo, or a mix with some other Stargell's card....this one is hard to follow, I must say.

Marc: Can you clarify what you mean by "agree on 50 of the 70 players"? Are you agreeing that these players are overpriced, or are you agreeing that these players are overpriced to the level that Riggio thinks they are? Based on my rating, the majority of the 70 on Riggio's list need an adjustment of +/- 25. So, most of them I'd consider "loosely priced correctly with only a minor adjustment needed".

Some of the adjustments are just head-scratchers. The 2006 O.Vizquel is one of the better SS in the entire set based on my ratings. If his price comes down by 1.1m I will be drafting him first in every single league. For point of reference, my rating say that 2006 O.Vizquel and 2009 J.Bartlett should be the exact same salary. From recollection, the Jet ratings loves the 2009 J.Bartlett, which probably means you guys do too. I am curious to know what your price differential, and what Riggio's price differential is between these two boys. My ratings say both of these boys should be priced within 0.05m of one another.

Priced just okay when used primarily at a secondary position
6.80M Pennington, when used as 3b
6.00M Stargell, when used as 1b---the 2.39M overestimation by Riggo must be a typo, or a mix with some other Stargell's card....this one is hard to follow, I must say.

Relative to other 3B, Pennington isn't that bad (because almost all 3B are overpriced). If you give a premium for switchhitters, that would explain why your ratings like him better.

Regarding '66 Willie Stargell, you are right, I have him priced as an OF. At 1B or DH, he's not so bad (but still worst value among his 4 cards).

Regarding '66 Willie Stargell, you are right, I have him priced as an OF. At 1B or DH, he's not so bad (but still worst value among his 4 cards).

I agree with this.

Can you clarify what you mean by "agree on 50 of the 70 players"? Are you agreeing that these players are overpriced, or are you agreeing that these players are overpriced to the level that Riggio thinks they are?

It's a mix of both. But I would say most the players identified by Riggo are priced in my rating system (based on neutral stadiium) 0.75M or worst than their salaries. And Riggo has on his list the most dreadful salaries I have on my own ratings, Luis Gonzalez, Guerrero's 9.53M, etc.

This said, following Cristano's suggestion, I reassessed the players based on the most fitted stadium, rather than based on neutral stadiums, and it does have a big impact on the salary assessment.

One extreme example Jose Bautista. Based on a neutral stadium, I have him rated at 6.68M, so roughly 1.3M below his actual salary. But Bautista is one of the rare player with zero ballpark singles and 8 ballpark homeruns. So when I adjust the ratings for home games being played at Polo Grounds (0=single 20=homepark), his value gets pushed all the way up to 8.02M, so pretty much on par with his salary. He could even become a good buy in the particular case where his capability of playing both 3b and rf saves you from getting a bench player covering third base. Perhaps this is why Riggo has me on record saying that I had some love for Bautista, even though my ratings have him rated almost as bad as his.

Thus, when ratings are based on the most optimal stadium, the high-priced players with lots of power whom Riggo identified as egregisouly overpriced are not bad as they look. Same with players with no power at all. The five worst rated players in my own ratings, after adjusting each card to its most optimal stadium, have the same profile: some pop, but not overpowering:

7.13M Bib Falk
9.45M Paul Waner
5.54M Paul O'Neil
7.04M Don Hurst
5.67M Spud Davis

With regards to 6.26M Vizquel.

Initially, my ratings had him at 5.45M in neutral stadiums, but since I give a premium of 0.45M to switch hitters, his cards gets rated 5.90M.

Adjusted to his most optimal stadium, his card gets a 0.51M bump to 6.41M. This most optimal stadium is based on 81 home games at Forbes and 81 away games in neutral stadiums. His bump would be further up assuming full time in Forbes Field.

This said, I recently downgraded the value of middle defense. This is based on an analysis of 30 Strat season of Napoleon Lajoie. When I analysed the data from Strat, his defensive performance didn't fit my model. I came to realize that Hal does a lot more than previously thought in my model of pulling infielders in or holding players. These strategies have the net result of downgrading the range of middle infielders (and first basemen). Since Vizquel is a 1e4, he is one of the most affected by my change of value. I downgraded his value by 0.47M. So I currently have him at 5.43M for neutral stadium, and 5.94M in Forbes park.

Perhaps my adjustment is too broad. I am still fine-tuning the process.

Bartlett's 6.68M card is also affected by this new adjustment, but to a lesser extent (downsize of 0.31M). I have him at 7.26M, and a bit higher in Forbes.

because almost all 3B are overpriced

I have to say, I was clueless when I read this, because positional adjustments are somewhat arbitrary: you just need to use what makes sense. If one position is systemically overpriced, you just need to change your position adjustment, and there is no other way since there is no "objective" way to define how one defensive position relates to another.

Here is a quote from the Society of American Baseball Research (my italics):

The values used for the position adjustments we use currently [...] were agreed upon and are the ones used by our sister site, FanGraphs:
Catcher: +12.5 runs (all are per 162 defensive games)
Shortstop: +7.5 runs
Second Base: +2.5 runs
Third Base: +2.5 runs
Center Field: +2.5 runs
Right Field: -7.5 runs
Left Field: -7.5 runs
First Base: -12.5 runs
Designated Hitter: -17.5 runs

We could agree or disagree with these adjustments. Personally, I would have adjusted differently left fielders and right fielders since the quality of arm has more impact in rf than in lf. But in any case, the point is to show that even the best sabermetricians didn't know how to formulate the positional adjustment: they used something that made sense. They used a scale so that the best shortstops are similar on the WAR scale compared to the best outfielders.

Doing the positional adjustments for Strat is different since you have to integrate the X-chart to the adjustment. But the end point is not far from the values above. Here is the average adjustment I have in my own ratings per position (this is actually an integration of positional adjustment and defensive value as generated by the defensive ratings--defensive value set at zero in the case of dh)

Catcher: +2 runs (mine are based on 150 defensive games)
Shortstop: +7 runs
Second Base: +5 runs
Third Base: +2 runs
Center Field: +4 runs
Right Field: -1 run
Left Field: -4 runs
First Base: -4 runs
Designated Hitter: -10 runs

The fact that the two positional adjustments are similar in scale is not surprising because SOM didn't want Omar Vizquel being priced higher than Ted Williams on the mere fact that Vizquel gets 170 more outs during the course of a season on the X-chart. We all know Ted Williams was the better player, and so the salary structure must reflect this. And therefore, the positional adjustment SOM used (deliberately or not) had to reflect this.

The catcher position shows the biggest difference between the agreed set of positional adjustments by sabermetricians and the ones used (as I estimate it) in SOM. And I think SOM got it wrong here. Cochrane and Yogi Berra were considered among the top 10 best players of their era, together they collected 5 MVP, but somehow their best cards are valued in the 6M-7M zone??? Something is wrong here. Had SOM used the positional adjustment agreed upon by sabermetricians, both Berra and Cochrane would have cards in the 8M zone at minimum.

Of course, I set my own positional adjustments so that I don't end up with massing overpriced or underpriced cards at any position. Had I used the 12.5 catcher adjustment, I would have a ton of underpriced cards, so that's why I settled on only +2 runs.

But if SOM is about to rescale the pricing structure, I would definitively suggest that it uses a positional adjustment so that the best catchers are more in line with how the baseball community perceived them.

Just want to come back to the gkhd11's team and my WAR analysis.

I showed that the Babe's 13.07M card performed, in his season played at Forbes Field, like a 11.25M card. This is within a few decimals of what my model predicts.

My model also predicts that the best Babe card is the 16.47M card. In a high-scoring environment, his card is expected over 13 WAR and be worth over 17M. This doesn't make Ruth as a super-bargain, you can get more from Bonds, among many others. But still, why is this card snubbed for the less effective Babe's 13.03M?

I think the answer to your query is plainly salary cap implications and the extra $3.5 million needed for the more expensive Ruth card. In addition, as with the Ted Williams cards for the most part, Strat does a terrible job of designing cards for players with super-high walks and super-high BA's. Ruth's best cards have far too few hits on them. Makes them almost always underperform in the more competitive leagues.

Just want to come back to the gkhd11's team and my WAR analysis.

I showed that the Babe's 13.07M card performed, in his season played at Forbes Field, like a 11.25M card. This is within a few decimals of what my model predicts.

My model also predicts that the best Babe card is the 16.47M card. In a high-scoring environment, his card is expected over 13 WAR and be worth over 17M. This doesn't make Ruth as a super-bargain, you can get more from Bonds, among many others. But still, why is this card snubbed for the less effective Babe's 13.03M?

Because your model, while it gives you numbers on a stand alone basis, does not account the card as actually used and is ultimately put to. The 13.07 card has 292/2160 natural HR chances on the left side and 240/2160 against righties, Babe was chosen for his ability to hit home runs on the road against left pitchers in a lefty park while having many natural HR’s against righties to disrupt righty pitchers as well hopefully to hit a home run and fatigue the pitcher so lesser batters are able to outperform, note the performance of Adams as a direct result of this. The 16.47 card has only 166/2160 natural HR’s and 160/2160 against righties and makes up for it walks and doubles, which would not be that helpful to my team, even if they were the same price. What Babe does when combined with a McGwire type hitter is add in value at Forbes when I can get the other players I managed to obtain, not subtract as your ratings tend to suggest. This is why you never see Ruth’s 13.07 card played as everyone is tied down to a single value Sabermetric rating when actual ratings are far more fluid than these ratings suggest.

I'm using the Babe's 16.47M card for the first time, and he's fairly awesome after 120 games: .356, .496, .746, 1.242 OPS. An all-around card in many ways. With 18 steals, I'm hoping he gets to 30 and can be a 30-30 Babe. He's leading off, too, because I want to maximize the number of plate appearances for my best hitters. https://365.strat-o-matic.com/team/1535402