After reaching 3 star On An Island on several of my players I've been watching for it in games and also had defensive schemes in which would try and leave these players in single coverage most of the time.
The Player
CB - Jordan Foles http://glb2.warriorgeneral.com/game/player/18725
Has 3* On An Island, 71 man cov (at time of sims), 70 coverage tech, 70 deflection, 40 vert and 30 diving.
Hypothesis
3* OaI has a 75% chance of firing; we'll assume this is if you have 100 man cov awareness so we'll say that the actual chance for Jordan is 75%*0.71 = 53.23% chance of firing whenever the conditions are met.
If we go through the last 2 games played by the Bogans we have 11 plays where it's clear the firing conditions are met (a handful more where it's debatable so I've excluded those, not that OaI fires on those either!)
Results
vs Faber Animals (League):
http://glb2.warriorgeneral.com/game/replay/36769/2723924
http://glb2.warriorgeneral.com/game/replay/36769/2723956
http://glb2.warriorgeneral.com/game/replay/36769/2724004
http://glb2.warriorgeneral.com/game/replay/36769/2724500
http://glb2.warriorgeneral.com/game/replay/36769/2724536
http://glb2.warriorgeneral.com/game/replay/36769/2724549
http://glb2.warriorgeneral.com/game/replay/36769/2725373
http://glb2.warriorgeneral.com/game/replay/36769/2725427
vs Las Vegas Outlaws
http://glb2.warriorgeneral.com/game/replay/50950/2671456
http://glb2.warriorgeneral.com/game/replay/50950/2672041
http://glb2.warriorgeneral.com/game/replay/50950/2672438
OaI doesn't fire once in 11 clear triggering situations.
Discussion
If we are to assume that our original hypothesis of 53.25% chance of firing and assume that all of these plays produced a roll for the SA then there is a 0.0233% chance of this occuring. I'm going to continue to collect plays for Jordan and may go through my other 2/3* OaI players (2*: http://glb2.warriorgeneral.com/game/player/21078, http://glb2.warriorgeneral.com/game/player/21079, http://glb2.warriorgeneral.com/game/player/21088. 3*: http://glb2.warriorgeneral.com/game/player/21081, http://glb2.warriorgeneral.com/game/player/21234)
I think it's clear that something in my assumptions is wrong so there are a few possible options as to what's causing these results:
1. There is no roll for OaI in these plays. This is because either I'm not interpreting the conditions correctly or there is a significant bug in the sim which doesn't trigger a roll under certain common conditions. Maybe the way it checks for a ball entering deflection range requires the defender to pass some additional awareness check?
2. The estimated firing percentage of 53.25% is wrong. This number is probably incorrect but I expect it to be within the ball park; I could work out if the method is correct on the more common SAs such as 3rd Down Stopper. Another possible bug is that there is an extra or missing 0 somewhere in the code which causes this to be a 5.325% chance or similar in which case the observed results would be entirely expected.
Conclusion
I think it's clear there is something fucked up about OaI, i'll continue to add results as games are played.
The Player
CB - Jordan Foles http://glb2.warriorgeneral.com/game/player/18725
Has 3* On An Island, 71 man cov (at time of sims), 70 coverage tech, 70 deflection, 40 vert and 30 diving.
Hypothesis
3* OaI has a 75% chance of firing; we'll assume this is if you have 100 man cov awareness so we'll say that the actual chance for Jordan is 75%*0.71 = 53.23% chance of firing whenever the conditions are met.
If we go through the last 2 games played by the Bogans we have 11 plays where it's clear the firing conditions are met (a handful more where it's debatable so I've excluded those, not that OaI fires on those either!)
Results
vs Faber Animals (League):
http://glb2.warriorgeneral.com/game/replay/36769/2723924
http://glb2.warriorgeneral.com/game/replay/36769/2723956
http://glb2.warriorgeneral.com/game/replay/36769/2724004
http://glb2.warriorgeneral.com/game/replay/36769/2724500
http://glb2.warriorgeneral.com/game/replay/36769/2724536
http://glb2.warriorgeneral.com/game/replay/36769/2724549
http://glb2.warriorgeneral.com/game/replay/36769/2725373
http://glb2.warriorgeneral.com/game/replay/36769/2725427
vs Las Vegas Outlaws
http://glb2.warriorgeneral.com/game/replay/50950/2671456
http://glb2.warriorgeneral.com/game/replay/50950/2672041
http://glb2.warriorgeneral.com/game/replay/50950/2672438
OaI doesn't fire once in 11 clear triggering situations.
Discussion
If we are to assume that our original hypothesis of 53.25% chance of firing and assume that all of these plays produced a roll for the SA then there is a 0.0233% chance of this occuring. I'm going to continue to collect plays for Jordan and may go through my other 2/3* OaI players (2*: http://glb2.warriorgeneral.com/game/player/21078, http://glb2.warriorgeneral.com/game/player/21079, http://glb2.warriorgeneral.com/game/player/21088. 3*: http://glb2.warriorgeneral.com/game/player/21081, http://glb2.warriorgeneral.com/game/player/21234)
I think it's clear that something in my assumptions is wrong so there are a few possible options as to what's causing these results:
1. There is no roll for OaI in these plays. This is because either I'm not interpreting the conditions correctly or there is a significant bug in the sim which doesn't trigger a roll under certain common conditions. Maybe the way it checks for a ball entering deflection range requires the defender to pass some additional awareness check?
2. The estimated firing percentage of 53.25% is wrong. This number is probably incorrect but I expect it to be within the ball park; I could work out if the method is correct on the more common SAs such as 3rd Down Stopper. Another possible bug is that there is an extra or missing 0 somewhere in the code which causes this to be a 5.325% chance or similar in which case the observed results would be entirely expected.
Conclusion
I think it's clear there is something fucked up about OaI, i'll continue to add results as games are played.






























