Live Update seems wrong, no I'm not just upset.

[BennytheKid]

Greinke gave up 7 HRs and gained 14 points of pHR, which was about 65% worse than his ZiPS projection for the year for HRA%

Jose Urena is leading the league in HRA with 9 and gained 14 points of pHR overshooting his ZiPS by 208% for HRA%

Connor Joe lost 8 points of Power while overshooting his ZIPs HR% this month by 75%. He also out preformed his ZiPS babip by 25% and lost 9 points of babip lastly he struck out more than the projection just barely at 6% more and gained 5 points of avoid k

It almost looks like a lot of these changes went the wrong direction? It doesn't seem this way on every card but their are dozens of confusing cards.

[BennytheKid]

Andre Jackson gained 7 points of pBABIP (weirdly 8 points vR and vL) and his babip was higher than ZiPS
He was the biggest gainer of pBABIP, my projections would've had him losing some of his rating
He also gained 3 pHR giving up 5 HR in 11.1 innings EDIT: I noticed he swapped to a RP from a SP so disregard this one.

Tim Hill gave up 2 HR in 14 innings which is worse than his ZiPS by 60% and gained 26 points of pHR

[kq76]

Connor Joe was the one that surprised me the most. I figured he'd surely go up and yet he went down -8? I only had 2 of him so I didn't get burned or anything. I was just surprised and hoped to learn why he dropped.

[BennytheKid]

Originally Posted by kq76

Connor Joe was the one that surprised me the most. I figured he'd surely go up and yet he went down -8? I only had 2 of him so I didn't get burned or anything. I was just surprised and hoped to learn why he dropped.

His ratings moved in the opposite direction of his actual performance relative to zips. He’s one of the very strange ones.

[evilsmokeybear]

It would be fantastic to get a little communication on this. We're not asking for a peek behind the curtain just some confirmation that it either did or didn't go as planned and if we can expect it to act the same in the future. Thanks

[eygabma]

BUMP - Can we get any sort of official response here??

[Will_L]

I seem to recall a pretty good discussion of this in last years version that had several good responses from the developers.

[bailey]

Originally Posted by Will_L

I seem to recall a pretty good discussion of this in last years version that had several good responses from the developers.

It would be nice if they had a policy of communication with "the greatest community in online gaming" (that's according to Rich Grisham) instead of taking that "greatest community" endlessly for granted.

[dking395]

Bumping this yet again.... OOTP Devs? Anyone pay attention to the community other than in live twitch streams?

[evilsmokeybear]

Originally Posted by Will_L

I seem to recall a pretty good discussion of this in last years version that had several good responses from the developers.

You're right. Looking at PT23 it does appear Matt Arnold was responding to questions about the Live Update. I wonder why the communication pattern has changed. Is there a way to ping an individual in the forum?

[Matt Arnold]

The baseline ratings are not purely based on zips. Some players will differ from there. As they play more, their season ratings will have a bigger impact and they should stabilize towards their season ratings.

[isidafo]

Originally Posted by Matt Arnold

The baseline ratings are not purely based on zips. Some players will differ from there. As they play more, their season ratings will have a bigger impact and they should stabilize towards their season ratings.

Thanks for the response Matt, I'm not sure it fully answers the question however. The way I had understood live updates to work is it would compare a player's real performance to a projection for that player (whether that's zips, steamer, something proprietary or a combination), and component ratings (and therefore overall) would increase if the real performance was better and decrease if it was worse. As far as I can tell that appears to align with what you've said here. However, there were a number of players that did not appear to behave that way. Taking Connor Joe as an example, he had a 159 wRC+ in April (career 100) and career highs in nearly all aspects. So, I have a difficult time believing that he underperformed his projections (whatever they're based on) by so much that he would go down by as much as any other player in this update. Is that the case?

[BennytheKid]

Originally Posted by Matt Arnold

The baseline ratings are not purely based on zips. Some players will differ from there. As they play more, their season ratings will have a bigger impact and they should stabilize towards their season ratings.

This does not address any of the examples provided. Jose Urena gave up 9 home runs in 18 IP and gained 14 points of pHR, Connor Joe is a similar example as Isidafo points out. There are obvious issues with this update that we could really use some clarity on.

[bailey]

Originally Posted by BennytheKid

This does not address any of the examples provided. Jose Urena gave up 9 home runs in 18 IP and gained 14 points of pHR, Connor Joe is a similar example as Isidafo points out. There are obvious issues with this update that we could really use some clarity on.

Your bug posts tend to be vague. If you want clarity you have to be clear yourself. If I recall correctly from your stream, this is what you wanted to know:

Did the team find issues with the update that will be fixed in the next update or is the update working as intended?

[BennytheKid]

Originally Posted by bailey

Your bug posts tend to be vague. If you want clarity you have to be clear yourself. If I recall correctly from your stream, this is what you wanted to know:

Did the team find issues with the update that will be fixed in the next update or is the update working as intended?

There is either a bug or ratings were changed a separate time from the games launch or live updates do not have anything to do with player stats anymore. Either way you are correct that what people need to know is what to do with cards like Connor Joe come the update in June. No functioning projection system zips or otherwise creates all the examples that have been given without an issue or something else changing.

Considering this has massive in-game economy effects it would be nice to know if there will be corrections to the broken system, changes to the affected players and or bugs being fixed.

[kq76]

I can't help but think it was a correction of a previous error. Anyone who knows anything about OOTP and looks at Connor Joe's 2022 statline couldn't possibly have pegged him as starting the year as a 78. He's a 1B/Corner OFer with, at that time, a sub 100 OPS+. How in the world could he possibly be rated a 78? It doesn't make sense. Him being rated a 70, after the update however, does. I haven't done an analysis of all the players like some probably have, but my guess is if someone looked back at the ones we're surprised by that most, if not all, of those players were probably over-rated from the start.

Ultimately, I think it's best to try not to take all this too seriously. It's unfortunate for those of us who like to predict how ratings will change as who knows what will happen next, but we have to keep in mind that whoever is coming up with the ratings is human and as such they'll have biases and make mistakes. This isn't a real stock market with billions of dollars at stake, it's just a silly little game only loosely based on baseball.

[BennytheKid]

Originally Posted by kq76

I can't help but think it was a correction of a previous error. Anyone who knows anything about OOTP and looks at Connor Joe's 2022 statline couldn't possibly have pegged him as starting the year as a 78. He's a 1B/Corner OFer with, at that time, a sub 100 OPS+. How in the world could he possibly be rated a 78? It doesn't make sense. Him being rated a 70, after the update however, does. I haven't done an analysis of all the players like some probably have, but my guess is if someone looked back at the ones we're surprised by that most, if not all, of those players were probably over-rated from the start.

Ultimately, I think it's best to try not to take all this too seriously. It's unfortunate for those of us who like to predict how ratings will change as who knows what will happen next, but we have to keep in mind that whoever is coming up with the ratings is human and as such they'll have biases and make mistakes. This isn't a real stock market with billions of dollars at stake, it's just a silly little game only loosely based on baseball.

I think this is a more than reasonable take for Connor's specific situation, I'm not sure it answers any of the others especially Urena. But it would be great to hear what the issue was and if we can trust the updates moving forward.

I'm pretty sure card ratings are not handmade as there is over a thousand live cards. Just asking for some answers on the why it was wrong and what can we expecting going forward. I think with millions of PP involved a heads up if things like Connor were going to happen would be greatly appreciated

[kq76]

Originally Posted by BennytheKid

I'm pretty sure card ratings are not handmade as there is over a thousand live cards.

I didn't mean that. I meant that whatever they programmed, they probably made a mistake. For example, I use spreadsheets a lot and I'm pretty good about not making mistakes with my formulas or at least catching them soon after, but occasionally I miss something, and depending on what the formulas are doing, the formulas could make most of the output look fine and only some of it would look obviously wrong.

With rating PT players, however they're doing it, they're probably weighting different ratings differently and if say a player has, for example, a BABIP that really stands out while most don't, if you accidently give BABIP a very different weight than you mean too, maybe even an inverse weight, well then any player who has a very different BABIP is going to look more wrong than players who have a pretty normal BABIP.

Who knows. I don't. I'm just guessing on all this.

[BennytheKid]

Originally Posted by kq76

I didn't mean that. I meant that whatever they programmed, they probably made a mistake. For example, I use spreadsheets a lot and I'm pretty good about not making mistakes with my formulas or at least catching them soon after, but occasionally I miss something, and depending on what the formulas are doing, the formulas could make most of the output look fine and only some of it would look obviously wrong.

With rating PT players, however they're doing it, they're probably weighting different ratings differently and if say a player has, for example, a BABIP that really stands out while most don't, if you accidently give BABIP a very different weight than you mean too, maybe even an inverse weight, well then any player who has a very different BABIP is going to look more wrong than players who have a pretty normal BABIP.

Who knows. I don't. I'm just guessing on all this.

Yup exactly why I am posting a bug report so they know they had an issue. I'm pretty well versed in spreadsheets myself so I completely agree here. Just looking to see if it is safe to invest moving forward and what went wrong/if it is fixed. I know a lot of people are talking about this one.

[BennytheKid]

3 Perfects to Diamond, No Diamonds to Perfect.

18 Diamonds to Gold, 6 Golds to Diamond.