Question about J.T. Realmuto's overall rating

[HondoLane]

I've really paid closer attention this year to the advanced stats and peripherals as I've tracked the live updates. And generally-speaking, I think I understand much better the process and logic you're using to determine player ratings, and it makes sense. However, there is one big outlier for me: J.T. Realmuto

I'll explain - the big discrepancy for me is where he started vs. his performance - meaning, if he started the season as a 90 overall, I could make sense of it; but he started the season as a 100.

As of the last update (not taking into account anything since September 4th), he already, at that point, had a career high in WAR - additionally, he was tied for his career high in DWAR (with a month left), while posting an also tied for his career-high 126 OPS+

As compared to last year, where he was a full 1 WAR less, in both overall and defense, with a 110 OPS+.

Starting the year at 100, and then being a 95 on September 4th with career numbers in both offense and defense... this one I really don't understand. What is J.T. Realmuto doing worse this year as opposed to last year?

Overall, the ratings seem to be accurate and fair. I'm just really not seeing how J.T. got his rating.

Thanks for taking the time to read and respond.

[HondoLane]

To add a second player whose ratings I am slightly baffled by: Devin Williams



Whether via advanced analytics or traditional stats, my man is one of the best relievers in baseball over the last three years - and this year. He has the 5th lowest FIP among relievers in 2022, with expected ERA of just 2.00 (only slightly above his actual 1.76 ERA). And yet... he's a gold card.

I understand his walk rate is high, but his insane strikeout rate and low HR rate clearly make up for it in the balance. As with Realmuto, is there any clarity you can offer as to why Williams has such a low rating in relation to his stats?

[Matt Arnold]

There's always some players the "algorithm" ends up liking more than others. Especially players who derive a lot of value from defense, stuff can be variable, and the live updates are likely not perfect at tracking his defensive value. So someone like Realmuto could easily swing by a lot if you tweaked his defense by a few points either way.

For Williams, the walks are very punishing to his overall. While often FIP is a good predictor for player ratings among pitchers (why someone like Gausman continues to rock high ratings despite his ball in play struggles), at the extremes, it doesn't necessarily work the same. The way the overall is calculated, it's penalizing him for the walk troubles more than FIP does because he's a fair bit below average there.

[HondoLane]

Originally Posted by Matt Arnold

There's always some players the "algorithm" ends up liking more than others. Especially players who derive a lot of value from defense, stuff can be variable, and the live updates are likely not perfect at tracking his defensive value. So someone like Realmuto could easily swing by a lot if you tweaked his defense by a few points either way.

For Williams, the walks are very punishing to his overall. While often FIP is a good predictor for player ratings among pitchers (why someone like Gausman continues to rock high ratings despite his ball in play struggles), at the extremes, it doesn't necessarily work the same. The way the overall is calculated, it's penalizing him for the walk troubles more than FIP does because he's a fair bit below average there.

Awesome - thanks for the explanation, Matt. I was trying to figure out if I was actually calculating something wrong-appears not.

[Tinkerman]

Realmuto is closing in on the second ever 20-20 season by a catcher, and hasn't been caught stealing yet. I'd love to know what his sliders are set to...

[HondoLane]

Originally Posted by Tinkerman

Realmuto is closing in on the second ever 20-20 season by a catcher, and hasn't been caught stealing yet. I'd love to know what his sliders are set to...

Whoa, I didn't even notice. Career-high OPS, career high in WAR (defensive, offensive and total) AND 18 stolen bases without getting caught? Beast year.

[HondoLane]

Originally Posted by Matt Arnold

The way the overall is calculated, it's penalizing him for the walk troubles more than FIP does because he's a fair bit below average there.

This is actually very revealing and explains most of the other disparities in player ratings that I couldn't quite account for. Thanks again, Matt.

[bear_rammedge]

You can work yourself into a tizzy on fangraphs trying to sort out the rankings, or you can just accept that it's continued market manipulation by the PT developers. Up to you.

[HondoLane]

Originally Posted by bear_rammedge

You can work yourself into a tizzy on fangraphs trying to sort out the rankings, or you can just accept that it's continued market manipulation by the PT developers. Up to you.

If you saw how much PP I've made on updates this year, you wouldn't say that, lol.

[bear_rammedge]

Originally Posted by HondoLane

If you saw how much PP I've made on updates this year, you wouldn't say that, lol.

Yeah, it's obviously still possible. But there have been so many headscratching updates this year (but not as many as last year since the game didn't launch until it was almost the all-star break,) combined with last year's "you can only have 10 of each card" rule that it's obvious what's going on. Every one of these games does it but OOTP is particularly egregious about it.

[HondoLane]

Originally Posted by bear_rammedge

Yeah, it's obviously still possible. But there have been so many headscratching updates this year (but not as many as last year since the game didn't launch until it was almost the all-star break,) combined with last year's "you can only have 10 of each card" rule that it's obvious what's going on. Every one of these games does it but OOTP is particularly egregious about it.

As far as the updates go, the more that I've honed in on what combination of metrics they are using (I think I'm close), the more all of them seem reasonable.

Realmuto and Williams really standout, but there aren't a lot of others where I don't already understand why they are rated so high or so low. And the explanation about Williams explains many other outliers.

I think the 10 card rule makes sense. Otherwise, I'd buy 50 of guys I know are going up, especially when the card is cheap. I mean, I bought 10 Hector Neris last month about a week before the update, all for under 1200. He went perfect (I didn't expect that, but I was 100% sure he was going diamond). I would have had 100 of him (he was one of the surest bets, but people just didn't know). If they dropped it to 5, I'd be irritated, lol.

To your last point: From my experience, it's the opposite - OOTP is the only game I can even think of where it is possible to bridge the gap between F2P and P2P.

[bear_rammedge]

The Show is by far the easiest to compete with P2P because ... there's no cap on card purchases and the roster updates are more frequent and actually make sense.

[HondoLane]

Originally Posted by bear_rammedge

The Show is by far the easiest to compete with P2P because ... there's no cap on card purchases and the roster updates are more frequent and actually make sense.

Never played it, so I couldn't make that comparison. But from all the games I have played, which are many (of the card-collecting/sports variety), OOTP is definitely the most competitive between F2P and P2P.

I've come to much prefer their method of rating players vs. using traditional stats. FIP actually IS a better indicator of performance for example, than the stats we're used to look at.

Out of curiosity, which player ratings do you find to be non-sensical at the moment? (aside from who I mentioned)

[bear_rammedge]

Originally Posted by HondoLane

Never played it, so I couldn't make that comparison. But from all the games I have played, which are many (of the card-collecting/sports variety), OOTP is definitely the most competitive between F2P and P2P.

I've come to much prefer their method of rating players vs. using traditional stats. FIP actually IS a better indicator of performance for example, than the stats we're used to look at.

Out of curiosity, which player ratings do you find to be non-sensical at the moment? (aside from who I mentioned)

The developers use the excuse of "we're going off the last three years, not just this year," which makes sense in a vacuum but it's really just an excuse to manipulate cards people are hoarding. Because they often use a poor 2022 performance to dock players who are historically great, but if a previously average or just good player explodes in 2022, they use 2020 and 2021 to dock them. You can't have it both ways (unless you're in charge and you want people to spend real money on perfect points and not get them for free by playing the auction house correctly.

For instance, Juan Soto is an 89 because his 2022 has been underwhelming. But if we're going off the last three years there's no reason to weigh his average (for him) 2022 so much more than his historically good 2020 and 2021.

Somehow though, his 4 fWAR 2022 has him ranked exactly the same as Gavin Lux, who has just 3 fWAR in 2022, was worth just 1 fWAR in 2021 (entirely due to baserunning) and had 69 plate appearances in 2020. There's no way to look at their 2022 or three-year resumes and conclude they're valued the same unless you think Juan Soto's previous 2,000 plate appearances were a mirage.

Trea Turner can play a better than average SS, 2B and CF. He's hitting an OPS over .800 in this brutal run environment after back to back >.900 OPS seasons. He's posting his second-best fWAR of his career. He's an 84, the same as Harrison Bader who hasn't posted a 4 fWAR season since 2018.

Nate Lowe is having a nice 2022, ninth-best fWAR among 1B (2.8) in his second-full season (1.7 fWAR in 2021.) He's an 89. In what universe is he valued the same as Austin Riley, who is wrapping up a >5 fWAR season after a 4.7 fWAR season and is a better hitter and plays a more valuable position?

Xander Bogaerts' 6 fWAR season (best of his career. in fact he's only been under 4 twice in the last 8 years, in 2017 when offense exploded (3.1 fWAR) and the covid-shortened 2020 when he was worth 1.6 fWAR in only 56 games) has him valued the same as David Fletcher, who has played 54 games and is worth 0.7 fWAR (after a 0.8 fWAR season last year in 157 games.)

Bogaerts is about to sign a $250-300m contract and David Fletcher is about to be non-tendered. But sure, yeah, they're the same.

These are just off the top of my head. Dylan Cease (4.4 fWAR last year, 4.6 this year) is an 80. AN 80! Ninth-highest SP WAR in baseball, but sure, he's ranked the same as Jeff Springs (2.6 fWAR in first full season) and Tarik Skubal.

Corbin Burnes was a 7.5 WAR pitcher last year with a sky-high K rate and miniscule walk rate. he was ranked 100 most if not all of 2021 on PT. He's struggled a bit in 2022, but he still has the 12th-best pitching WAR this season and his rate stats are only down a smidge. If anything his struggles are overblown. But sure, he's a 78.

There are dozens more of these, but you get the point. The developers can come up with whatever reasons they want for these ratings, but when their own explanations are not consistent, the only answer left is market manipulation.

[HondoLane]

I agree that people like Boegarts, Devers, Cease, Alcantara and a lot of others sort of start with a shaft by the way the ratings system works. But if you understand that is how it works, it's not hard to follow who is going up or down.

I mean this in the general sense that I have not seen a player's OPS rising and the card rating go down. So it may seem odd that Trea Turner dropped from a 92 to 84 while his average is similar, but his OPS dropped 30 points. If his OPS had risen 30 points, he'd be a 98. Also have not seen FIP drop significantly (with considerations) without the card going up, and vice versa on both. In this regard, the ratings can be predicted fairly.

[bear_rammedge]

Originally Posted by HondoLane

I agree that people like Boegarts, Devers, Cease, Alcantara and a lot of others sort of start with a shaft by the way the ratings system works. But if you understand that is how it works, it's not hard to follow who is going up or down.

I mean this in the general sense that I have not seen a player's OPS rising and the card rating go down. So it may seem odd that Trea Turner dropped from a 92 to 84 while his average is similar, but his OPS dropped 30 points. If his OPS had risen 30 points, he'd be a 98. Also have not seen FIP drop significantly (with considerations) without the card going up, and vice versa on both. In this regard, the ratings can be predicted fairly.

This is what I mean - you can fudge around with whatever explanation you want. But if your system looks at Trea Turner (for instance) and says “he’s an 84.” Your system is either wrong or you’re manipulating it on purpose.

[HondoLane]

Originally Posted by bear_rammedge

This is what I mean - you can fudge around with whatever explanation you want. But if your system looks at Trea Turner (for instance) and says “he’s an 84.” Your system is either wrong or you’re manipulating it on purpose.

I don't see that as fudging around - that's what I'm saying. It's consistent. Turner was an 88, his OPS went up 20 points, he went up to a 92. It went down 30 points, he dropped to an 84. That's easy to follow as long as you're following the right numbers.

The rating system itself definitely favors certain styles of play. I think that Trea Turner is rated so low because he is a SS, where he is poor defensively. If he was a 2nd basemen, he'd be rated a lot higher. A poor defensive SS with a plus bat seems to get a big penalty. Same thing goes for Bogaerts.

I think the best fix for this is to use a system like they have, and then make some manual changes for the players whose ratings clearly don't match their production (over some time).

[Tinkerman]

I've done quite well out of the live updates despite not actually buying any live cards. However, I'd swap all of those points for a couple of PD wins and pulling the number of perfects that I expect to. Would've been a lot more exciting than just more bean counting.

[bear_rammedge]

Originally Posted by HondoLane

I don't see that as fudging around - that's what I'm saying. It's consistent. Turner was an 88, his OPS went up 20 points, he went up to a 92. It went down 30 points, he dropped to an 84. That's easy to follow as long as you're following the right numbers.

The rating system itself definitely favors certain styles of play. I think that Trea Turner is rated so low because he is a SS, where he is poor defensively. If he was a 2nd basemen, he'd be rated a lot higher. A poor defensive SS with a plus bat seems to get a big penalty. Same thing goes for Bogaerts.

I think the best fix for this is to use a system like they have, and then make some manual changes for the players whose ratings clearly don't match their production (over some time).

this of course is not consistent with reality or how they even describe how their rating system operates. Of course you can look backward and retrofit whatever you want to decide that it makes sense that Trea Turner, one of the 10-15 most valuable baseball players in the league, is an 84 because his OPS dropped 30 points over a small sample of games. Or you can live in reality and acknowledge that makes no sense whatsoever.

[HondoLane]

Originally Posted by bear_rammedge

this of course is not consistent with reality or how they even describe how their rating system operates. Of course you can look backward and retrofit whatever you want to decide that it makes sense that Trea Turner, one of the 10-15 most valuable baseball players in the league, is an 84 because his OPS dropped 30 points over a small sample of games. Or you can live in reality and acknowledge that makes no sense whatsoever.

Sorry, man, you're the one living outside of reality on this one. I live in reality, and that's why I nail all the updates.

I've basically been right on every card for 3 months in a row. They are easy to track and follow. The updates make sense if you know what you're looking for.

Like I said, Turner's rating are very position dependent; he is a probably a 98 if he's a 2B. Yeah, that matters, and it's the way the base game works to. Nothing surprising about it. Really nothing to complain about. I've tried reasoning with you, lmao, but you don't concede any points, ever, regardless of what I say

Good luck, big guy