ATG 9 Too Flawed?

I am kneeling over ... egads! Of course, this SOM universe is not perfect. We deal in reality. Well put, FrankieT!

Fair point Hack. But there is a difference between imperfection and extreme disparities that go untouched forever.

I don't really understand the general points you are trying to make specific to ATG cards. For pricing purposes I work from a giant data sheet of baseball events and probabilities that reflect the outcomes on each card. these are then run through the ATG9 model first to determine a runs created value for each card and then to convert that value into a salary. Where the card originated from has no bearing on the pricing. Each of the added player nominations have all gone through this same filter as they have been added.

To your more specific question of justifying Wesley and Molina prices, I was going to do a complete mathematics breakdown of how they compare to others, but to FrankieT's point I can't see anyone waiting to slog through that analysis, so i will take a different approach in responding. If you were a reviewer and provided me with feedback that so and so is mispriced the first thing I would try to do is have you help me identify what factor is off. Any value or lever I play with in the model to adjust Wesley and Molina upward has unintended consequences on several thousand other cards. Lets assume for this exercise you can't identify the cause, but you just know something is off, I would then go break down the analytics to see what I can find. For this breakdown I just focused on 1B in the $650k to $550k range with similar defensive ratings to see who compares to Wesley (1B is his best defensive position).

When I do this analysis Wesley has a runs created value based on a constant number of PA's than players priced above him, consistent with your observation. However, not all runs are created equal, since baseball is governed by outs and not PA's. So the next step in the pricing model is to layer on the replacement value impacts of putting Wesley into your line up. So while Wesley may individually produce more runs he consumes significantly more outs in the process taking away PA's and runs from your better hitters. In ATG terms it is about .18 runs per incremental out consumed. When this is factored in Wesley loses any incremental value.

Now as a tester, you might disagree with the value assigned to these lost outs and that would be an assumption open to debate, but the concept is still accurate. Also you might agree with the overall concept but disagree that below $1M that is not how players are used and pricing should be focused more on a specific attribute instead of the entire card. That could also be an appropriate solution to refine pricing at this level, but again it is a refinement of an assumption. the current values assume the same number of PA's for all hitters, and incorporates all other aspects, defense, running, etc. when setting the pricing.

I didn't break apart Molina, but in just a glance at his card it is atrocious offensively. You might think that for a backup, that great D makes up for it and maybe you have found some value if you can use him defensively and avoid him coming to the plate. The fact he is used in all leagues doesn't make him priced incorrectly. ATG has an artificial cut off at $500k. By definition, the community is going to gravitate to the best players in this $500k pool to use as reserves. If the cut off gets drawn at a different point then different $500k guys become the ones always being used.

However, if you think that defense makes Molina a great bargain to start at $500k, then you can kiss about 100PA's from the rest of your line up good bye over the course of the season, which more than offsets any value you might think he has.

As others in this thread have pointed out the game is about finding value. If you can use someone in a park environement, L/R situation, defensive subs, pinch hitter role, etc. that maximizes there positives while limiting the negatives then you have found value. The pricing incorporates all of the factors based on baseline assumptions and it is the job of good managers to exploit the environment compared to these baselines.

And I would like to apologize to everyone else for making a thread that long to discuss the merits of Wesley and Molina.

Great post childsmwc.
And a blast from a dear past re: a potential (maybe problematic) way to do dynamic pricing that is scoped but I still don't think it would ever be done. The bigger point is the wisdom in it.

Pricing also does not need to be absolutely precise, discovering good values is an integral part of the game. However, when the same players become critical autodraft priorities, particularly the guys mentioned above at lower caps, then the game suffers and becomes something no one intended it to be - no one with half a brain at any rate.

There are a couple of significant problems with usage based models. A lot of the top players may well be reasonably priced but are popular just because they are real all time greats. Think Ruth, Mays, Williams etc. Usage based pricing would unfairly impact these guys. But the bigger obstacle is getting Strat to do the work to create usage based pricing. No way that happens with Richman still at the helm. That is why I proposed changing the pricing on 25 players each time players get added to ATG. It is a simple, quick and dirty fix with long reaching positive repercussions for the integrity of the game without being labor intensive for SOM and thus cost prohibitive.

What he said. Exactly.

I am not sure why you're apologizing as I think it's great to get a look at the methodology. Thank you. What you said sounds pretty good to me. I agree with much of what you said. In the case of 500k players they are quite popular/optimal as backups and generally speaking they're not very impressive offensively so while your point is valid about them costing your team plate appearances I would argue that this is the norm until the salary starts to increase.

I still believe Molina and Wesley are superior as many of the 500k players are quite lackluster (and rightfully so). I do think it's fun to hunt for hidden gems and there are many good examples of well balanced but useful 500k players. I think things start to break down quickly when certain cards are compared. I'm not sure why it is the case but many cards have few if any redeeming qualities. For example: Let's compare Bengie Molina to clarkson Brazelton.

Molina wins in: defense rating, arm, t-split, passed balls, fewer strikeouts (none), singles, doubles, homeruns, OBP, BA
Brazelton wins in: double plays (impactful),clutch

I don't see why anyone would ever take Brazelton unless it was for a theme league or a favorite player or something like that. He seems completely atrocious to me. There are lots of cards like that in the game and this is why I feel more updates with pricing would be good. In the case of 500k players I assume you're not going to enhance their card values so I feel it would be best to either remove them or replace them with other options. Molina IMO should cost more than 500k if cards like Brazelton cost 500k. Without looking I am confident it would be easy to find 550k+ cards that are weaker than Molina by a significant margin.

I could look at Wesley if you want but I've been rightfully accused of lost posts Instead, I'll just suggest you look at Ed Goodson. It doesn't get much worse than that!

For example: Let's compare Bengie Molina to clarkson Brazelton.

I can't understand what you're even on about at this point and I'm only replying because you DM'ed me asking me to weigh in.

You are comparing two 50 cent hitters. No one disputes that Molina is one of the best 50 cent catchers and more valuable than Brazelton. They have the same salary because 50 CENTS IS THE MINIMUM and neither can be reduced any further. What in god's name is the problem? Someone has to be the best 50-center, why not Molina? (I don't even have Molina as the best 50-cent catcher, FWIW.) Wesley is the best 50-cent first baseman, what of it? There are still 9 50-centers at other positions that I have as more valuable. Neither card is manifestly mis-priced at 50 cents, and the existence of inferior 50-cent cards certainly does not suggest as much, because again, 50 CENTS IS THE MINIMUM. By definition, the quality of 50 centers is going to range from the very worst cards up to replacement level. In a neutral $80 environment there are only 13 50-cent hitter cards above replacement level, exactly one of which is worth more than 0.3 WAR, out of 200 total cards. That's a pretty damn good job by the pricing committee if you ask me!

For example: Let's compare Bengie Molina to clarkson Brazelton.

I can't understand what you're even on about at this point and I'm only replying because you DM'ed me asking me to weigh in.

You are comparing two 50 cent hitters. No one disputes that Molina is one of the best 50 cent catchers and more valuable than Brazelton. They have the same salary because 50 CENTS IS THE MINIMUM and neither can be reduced any further. What in god's name is the problem? Someone has to be the best 50-center, why not Molina? (I don't even have Molina as the best 50-cent catcher, FWIW.) Wesley is the best 50-cent first baseman, what of it? There are still 9 50-centers at other positions that I have as more valuable. Neither card is manifestly mis-priced at 50 cents, and the existence of inferior 50-cent cards certainly does not suggest as much, because again, 50 CENTS IS THE MINIMUM. By definition, the quality of 50 centers is going to range from the very worst cards up to replacement level. In a neutral $80 environment there are only 13 50-cent hitter cards above replacement level, exactly one of which is worth more than 0.3 WAR, out of 200 total cards. That's a pretty damn good job by the pricing committee if you ask me!

How can you not understand it? I'ts very simple: A LOT of cards are horrendous. What is the point of putting them in the game? Especially in "All Time Greats" when they apparently don't know how to hit the baseball. One of the guys I just mentioned literally has 0 hits against lefties and barely any against righties. How is that a "damn good job"? I've never seen anything like this is any other game in my entire life. One would think that the point of a card existing is for it to be played.

Good god man, the pricing committee didn't put the cards in the game, they just priced them, and they did so accurately and reasonably. I don't know why Strat wants to spend their time adding unplayable cards but the presence of those cards detracts exactly zero from my enjoyment of the game. I don't use them! Simple as!

Forgive me for not knowing all the processes of how cards enter the game. All I am saying is the prices do not match the actual values of the cards in certain extreme cases. All I'm asking for is an occasional balance sweep of extreme cards. I don' t think that is unreasonable at all! I wish it was done at least a couple times per year to help the meta evolve. A good example of the importance of this is the relievers - which just about everyone seems to think are overpriced in general (maybe not specialists). I'd personally love to get back to using 3-4M relieves in 80M but they're just not good values.

I come from a background of magic the gathering (overpowered cards get banned) and MMORPGS (overpowered characters get nerfed) and so on. I don't understand why in Strat the cards are just released and left as is potentially forever. I think repricing updates could be very good for the game and fun to digest. Balance updates are a major part of the excitement of online games and do a great job and bringing back bored players eager to try something fresh.