no content
AdBlocker active?
It seems you are using software to block advertisements. You could help us if you could switch it off when visiting redzoneaction.org.
The reason is very simple: Advertisements help us running the site, to offer you the game in a good quality for free. So if you like the game, please support us by purchasing a Supporter Account or disabling the AdBlocker on this site.
Thank you very much!
| Main / Discussions / game Search Forum | |
| Navigation: |< < 1 2 3 4 5 >| | |
| Poster | Message |
| posted: 2023-12-23 20:46:21 (ID: 100179709) Report Abuse | |
|
HH_KMN wrote:
Quimriera wrote:
another thing is trying to "dissect" the engine to know exactly how it works. I sincerely think it is a waste of time. 
Its definitely not a waste of time. Just don't expect to always win despite having a high percentage play and a better matchup. I don't think that there is disagreement here. There is general agreement that there is a point in doing some analysis because otherwise playbooks and player selection would be just random and we might as well roll dice. There is also general agreement is no point trying to get right down into the finer points of the engine because you will never get a sample size and a definition of every parameter that would allow you to reverse engineer the entire engine including the weighting of every parameter and any allowance for random variation. |
|
| Quote Reply Edit | |
| posted: 2023-12-23 20:53:46 (ID: 100179711) Report Abuse | |
|
JohnHW wrote:
There is ... no point trying to get right down into the finer points of the engine because you will never get a sample size and a definition of every parameter that would allow you to reverse engineer the entire engine including the weighting of every parameter and any allowance for random variation. I hope so. Quite some years ago there was a guy doing "Beckos". This was a statistical approach to guess the outcome of matches. And it was quite good in doing so. I have never seen anything similar good in breaking down what would happen in future games. |
|
| Quote Reply Edit | |
| posted: 2023-12-23 21:48:15 (ID: 100179715) Report Abuse | |
|
If I remember correctly, he only did Beckos before you put in ELO. I think they were functionally the same.
|
|
| Quote Reply Edit | |
| posted: 2023-12-24 09:25:59 (ID: 100179727) Report Abuse | |
|
He said he was using a different initialization on his ELO, with changed parameters. And it worked much better than my ELO
|
|
| Quote Reply Edit | |
| posted: 2024-02-06 08:19:36 (ID: 100181020) Report Abuse | |
|
jack6 wrote:
So the results are for pass-mid: on roughly 2200 pass attempts I got 46.6% completion rate with an average of 14.6 yards on a completed pass. But I got also 12.8% INT rate on those 2200 passes. Translated that means, if you are desperate to play pass-mid, your chance for a competed pass is under 50%, which is not bad, since it is almost for sure than a 1st down. But also every 8th pass is an INT. That looks more promissing than I did expect it. Will do the same with pass long at some point. So I did now made some scrimmage games with same setup, but all pass-long. THAT did turn out to be a point festival, which gave me the idea on another nuts-competition like LEATHERHEADS, but I'm drifting off .... On roughly 4000 pass attempts (i did setup more scrimmages than in the test before) I got 16,36% competion rate with an average of 23.55 yards on a completed pass. I got 16,42% INT rate on those 4000 passes. Translated that means, if you need a pass-long, your chance to hit the target is more or less the same as to hit the defender and the chance to hit nothing is double as both chances cobined. That's bad. 1/6 goes to the target and is a catch, 1/6 is an INT and 2/3 will be dropped or sacked. I guess to win a game with that play with more than 24 yards needed is very very slim. |
|
| Quote Reply Edit | |
| reply Mark this thread unread | |
| Navigation: |< < 1 2 3 4 5 >| | |
| Main / Discussions / game | |
