|
Post by squirrel on Sept 27, 2018 16:44:35 GMT -5
|
|
|
Post by thefatcat on Sept 27, 2018 16:45:22 GMT -5
|
|
|
Post by squirrel on Sept 27, 2018 16:46:41 GMT -5
Nick and Joel I see your logic and I've been somewhat down this track.
If you get this for all QBs in a league you can get the distribution of the hidden rating, whatever that means. Then you can work backwards from there to get something. I couldn't get anything more predictive out of doing that...doesn't mean there isn't something there
|
|
|
Post by Nick on Sept 27, 2018 16:48:41 GMT -5
I guess Danny is +7.2 then. Apparently my backup Jimmie Fulton has a +7.3. But he's not fully developed so who knows how accurate that is. Understand a +7 for him and Antonik are two entirely different things as Antonik is starting from a higher overall rating. Antonik has excellent AI. Fuller just decent/good.
|
|
|
Post by Nick on Sept 27, 2018 16:53:49 GMT -5
Rakkoon is coming out REALLY good. Seb used to force a lot of passes back in the day.
Perry looks pretty good, no surprise there. Hutchings also excellent.
|
|
|
Post by jeremy on Sept 27, 2018 16:55:10 GMT -5
|
|
|
Post by squirrel on Sept 27, 2018 16:55:36 GMT -5
Apparently my backup Jimmie Fulton has a +7.3. But he's not fully developed so who knows how accurate that is. Understand a +7 for him and Antonik are two entirely different things as Antonik is starting from a higher overall rating. Antonik has excellent AI. Fuller just decent/good. Not sure I agree with this, FWIW. So take 3rd down, for example. Contributes 14% to the overall rating. So a guy with a 50 bar gets 50 * 14.0% = 7.0 more overall rating because of that bar. That guy has a 60/60 rating. Right, next guy comes along, he has a 30/30 overall rating. He has a 50 bar for 3rd down also. And the contribution to overall rating from that bar...the same. 7.0 more added to the overall. My working assumption would be the same logic extends to avoid INT. The rating is hidden but the logic is the same. The 80/80 and the 20/20 get the same impact from the bar.
|
|
|
Post by ManWithNoName on Sept 27, 2018 16:57:57 GMT -5
Be interesting to see a list of all the starting QBs that have played at least a couple seasons so we could compare the rating to the stats but it seems like it's pretty accurate. I don't know if Danny's is really that high but I looked at Rico and I have him at -3.0 and he's always had trouble with interceptions. Thanks for sharing this, Squirrel.
|
|
|
Post by Joel on Sept 27, 2018 16:58:04 GMT -5
Nick and Joel I see your logic and I've been somewhat down this track. If you get this for all QBs in a league you can get the distribution of the hidden rating, whatever that means. Then you can work backwards from there to get something. I couldn't get anything more predictive out of doing that...doesn't mean there isn't something there So heres where I went once I found out the +/-, and feel free to call it dumb, because I am not sure it works, as it hits a snag on certain guys I think. So if we take Danny Antonik, Evan's QB, he is a 64 overall, but the bars before come to 56.8, meaning he has a +7.2. Which means the AI on him has a weight of 11.25%. 11.25 x 7.2 = 81 Therefore I would have his AI as 81.
|
|
|
Post by Bighouserulez on Sept 27, 2018 17:00:56 GMT -5
yep that gives my GML guy Scott +1.9 which is 186 and 79 in his career.
Shaka's all world guy Stevens gets a +8.8
|
|
|
Post by ManWithNoName on Sept 27, 2018 17:03:23 GMT -5
Nick and Joel I see your logic and I've been somewhat down this track. If you get this for all QBs in a league you can get the distribution of the hidden rating, whatever that means. Then you can work backwards from there to get something. I couldn't get anything more predictive out of doing that...doesn't mean there isn't something there So heres where I went once I found out the +/-, and feel free to call it dumb, because I am not sure it works, as it hits a snag on certain guys I think. So if we take Danny Antonik, Evan's QB, he is a 64 overall, but the bars before come to 56.8, meaning he has a +7.2. Which means the AI on him has a weight of 11.25%. 11.25 x 7.2 = 81 Therefore I would have his AI as 81. Assuming Avoid Interceptions is the only hidden bar being calculated that has any weight, makes perfect sense to me.
|
|
|
Post by Bighouserulez on Sept 27, 2018 17:03:57 GMT -5
|
|
|
Post by Bighouserulez on Sept 27, 2018 17:04:30 GMT -5
So heres where I went once I found out the +/-, and feel free to call it dumb, because I am not sure it works, as it hits a snag on certain guys I think. So if we take Danny Antonik, Evan's QB, he is a 64 overall, but the bars before come to 56.8, meaning he has a +7.2. Which means the AI on him has a weight of 11.25%. 11.25 x 7.2 = 81 Therefore I would have his AI as 81. Assuming Avoid Interceptions is the only hidden bar being calculated that has any weight, makes perfect sense to me. I wonder on the hidden fumble rating and it that comes into play for a QB here too.
|
|
|
Post by Joel on Sept 27, 2018 17:06:55 GMT -5
Assuming Avoid Interceptions is the only hidden bar being calculated that has any weight, makes perfect sense to me. I wonder on the hidden fumble rating and it that comes into play for a QB here too. <iframe width="27.28" height="4.84" id="MoatPxIOPT0_79917506" scrolling="no" style="border-style: none; left: 15px; top: -5px; width: 27.28px; height: 4.84px; position: absolute; z-index: -9999;"></iframe> <iframe width="27.28" height="4.84" id="MoatPxIOPT0_20258355" scrolling="no" style="border-style: none; left: 1300px; top: -5px; width: 27.28px; height: 4.84px; position: absolute; z-index: -9999;"></iframe> <iframe width="27.28" height="4.84" id="MoatPxIOPT0_79192389" scrolling="no" style="border-style: none; left: 15px; top: 181px; width: 27.28px; height: 4.84px; position: absolute; z-index: -9999;"></iframe> <iframe width="27.28" height="4.84" id="MoatPxIOPT0_60246825" scrolling="no" style="border-style: none; left: 1300px; top: 181px; width: 27.28px; height: 4.84px; position: absolute; z-index: -9999;"></iframe> I don't think it does, as it comes along with Scrambling which has a 0% weight, as does Timing.
|
|
|
Post by Nick on Sept 27, 2018 17:09:40 GMT -5
Understand a +7 for him and Antonik are two entirely different things as Antonik is starting from a higher overall rating. Antonik has excellent AI. Fuller just decent/good. Not sure I agree with this, FWIW. So take 3rd down, for example. Contributes 14% to the overall rating. So a guy with a 50 bar gets 50 * 14.0% = 7.0 more overall rating because of that bar. That guy has a 60/60 rating. Right, next guy comes along, he has a 30/30 overall rating. He has a 50 bar for 3rd down also. And the contribution to overall rating from that bar...the same. 7.0 more added to the overall. My working assumption would be the same logic extends to avoid INT. The rating is hidden but the logic is the same. The 80/80 and the 20/20 get the same impact from the bar. That doesn't hold up for me. Jeremy's Stafford couldn't possibly have a +7 as his rating doesn't allow for that high a positive. I am pretty sure these figures you are coming up are relative to the players overall.
|
|