wr overalls

[purplerat]

Quote: Originally Posted by Dempseylicious23

It doesn't contradict that at all. What I'm saying is, while OVR may be the most highly contributing factor in determining simulation stats, it is not necessarily the only defining factor that determines simulation stats. Overall ratings are based on your schemes, player position types, and the player's skills themselves. There should be a difference if you play Wes Welker in the slot as a possession receiver than if you played him out wide all the time as a red zone threat in both actual play and OVR rating (which is reflected in your depth chart in the menus). One could assume that Welker would perform far better in the former situation. Having a player do what they are used to doing in a system optimally set for their skill set should see better results.

What you fail to explain is how OVR would impact sim stats. Do you think it's a comparison like 90wr vs 85cb = wr wins? or that game stats are just randomized with overall being a function of what the outcome is? Both of those would be in contradiction to a sim system which actually looks at all factors. I don't see any sensible way in which OVR is used to determine sim results other than if you think the simming is a very rudimentary system like a card game or something. If it's a complex system that considers many factors there would be no reason to consider OVR since OVR is 100% derived from those other factors.

In other words you are saying that the sim system chooses to use a dependent variable in place of much more reliable independent variables which is just as easily has available. That would be a very dumb way to go about doing simulations.

[Dempseylicious23]

Quote: Originally Posted by purplerat

What you fail to explain is how OVR would impact sim stats. Do you think it's a comparison like 90wr vs 85cb = wr wins? or that game stats are just randomized with overall being a function of what the outcome is? Both of those would be in contradiction to a sim system which actually looks at all factors. I don't see any sensible way in which OVR is used to determine sim results other than if you think the simming is a very rudimentary system like a card game or something. If it's a complex system that considers many factors there would be no reason to consider OVR since OVR is 100% derived from those other factors.

I fail to explain it because no one outside EA's Madden division is going to know exactly how OVR impacts sim stats.

Look, you're insulting my intelligence at this point. Of course a situation where you have a 90 WR vs 85 CB doesn't just gift wrap the WR a win every time. That would be silly.

I also never said that OVR was the penultimate determining factor. I'm simply saying it COULD be. I'm not sure why you feel the need to rebut every point I make (well, not every point, you didn't feel the need to include my car analogy or my YPC vs YAC retort I see). I'm only purporting that it COULD be AWR, or it COULD be OVR, and that there is potentially a way to try to determine which is more important to simulation stats.

[purplerat]

Quote: Originally Posted by Dempseylicious23

I fail to explain it because no one outside EA's Madden division is going to know exactly how OVR impacts sim stats. Look, you're insulting my intelligence at this point. Of course a situation where you have a 90 WR vs 85 CB doesn't just gift wrap the WR a win every time. That would be silly. I also never said that OVR was the penultimate determining factor. I'm simply saying it COULD be. I'm not sure why you feel the need to rebut every point I make (well, not every point, you didn't feel the need to include my car analogy or my YPC vs YAC retort I see). I'm only purporting that it COULD be AWR, or it COULD be OVR, and that there is potentially a way to try to determine which is more important to simulation stats.

I'm not expecting you to have the actual explanation on how it works, but if you think it factors in then it would only make sense to have at least some guess on how it may work. I fail to see any useful explanation of how OVR would practically be used in determining sims. Maybe this is because I have some experience programming simulations and understand why the Madden OVR would be a poor value to use in the type of simulation I believe is being run.

The point I'm trying to get at is that Overall ratings in Madden are 100% dependent variables and therefore have little to none utility in a complex simulation engine which has direct access to all of the independent variables. It would be a failing on the part of the developers if it were the case that OVR directly influenced the results of the simulation.

[Dempseylicious23]

Quote: Originally Posted by purplerat

I'm not expecting you to have the actual explanation on how it works, but if you think it factors in then it would only make sense to have at least some guess on how it may work. I fail to see any useful explanation of how OVR would practically be used in determining sims. Maybe this is because I have some experience programming simulations and understand why the Madden OVR would be a poor value to use in the type of simulation I believe is being run. The point I'm trying to get at is that Overall ratings in Madden are 100% dependent variables and therefore have little to none utility in a complex simulation engine which has direct access to all of the independent variables. It would be a failing on the part of the developers if it were the case that OVR directly influenced the results of the simulation.

From my viewpoint, you make a good case with this post. A good programming squad would implement a lot of variables that factor in to the simulation stats engine. However, Adembroski's recent twitter rants seem to indicate that his work on the simulation stats engine was undone by someone in a higher position than himself, leading to some of the wacky stats that happen during CCM's over time. If what Adembroski ranted about was true, it leads me to believe that his work was good and took into account a lot of variables, but may have been unpolished and buggy in certain instances which were not conducive to what the brass wanted (or maybe not, it could have been inter-company politics or something way deeper than that).

Then it would make sense that a lead designer might rework the code at a later time (and now we're on the hook to meet deadlines and such) to a much simpler system. That's why it's also very possible that OVR is a large determinate factor in the equations.

Good programming leads one to take your viewpoint. Bad programming by one of the worst managed/run corporations in America done in a short time frame leads one to think that the system may not be as complex as you believe.


Edit: I have a degree in Mechanical Engineering with a background versed in Matlab and Mathcad programming and trust me, I know how critical it is to be proactive about writing your scripts. If you do something that looks great but just has one piece missing that you can't determine in hundreds of lines of code, you need to scrap it and move on to something simpler that you can complete within the remaining time you have.

[purplerat]

Quote: Originally Posted by Dempseylicious23

From my viewpoint, you make a good case with this post. A good programming squad would implement a lot of variables that factor in to the simulation stats engine. However, Adembroski's recent twitter rants seem to indicate that his work on the simulation stats engine was undone by someone in a higher position than himself, leading to some of the wacky stats that happen during CCM's over time. If what Adembroski ranted about was true, it leads me to believe that his work was good and took into account a lot of variables, but may have been unpolished and buggy in certain instances which were not conducive to what the brass wanted (or maybe not, it could have been inter-company politics or something way deeper than that). Then it would make sense that a lead designer might rework the code at a later time (and now we're on the hook to meet deadlines and such) to a much simpler system. That's why it's also very possible that OVR is a large determinate factor in the equations. Good programming leads one to take your viewpoint. Bad programming by one of the worst managed/run corporations in America done in a short time frame leads one to think that the system may not be as complex as you believe.

What you say about code being changed later and using lazy programming to meet a deadline could very well be the case. Trust me I know all about how that goes down.

One point of clarification though is that I would call it good/bad programming in relation to how the stats come out in terms of realism. The programing could be good and have unrealistic results as well could it be bad and have realistic results. So I wouldn't jump to wonky stats being the result of some major programming flaw such as using OVR ratings where they really shouldn't be used.

[Old School SD Fan]

Quote: Originally Posted by Dempseylicious23

I'd wager it's more the OVR rating that impacts simulation results. It's hard to tell though since OVR is so heavily weighted by AWR, so it's difficult to really say which is the more important factor here. I guess a way to check would be to get two WR's with the same OVR, one with high AWR and low physicals, one with high physicals and low AWR, and sim through a few seasons with both of them to see if there is any dramatic difference in their performance. That still might not really prove anything concrete though.

None of us know how sims are programmed and the heavy link between overall and AWR makes it tough to test to find out. It's just MHO that overall affects simming. Perhaps it's just a simple tool to help quickly judge who to start or to set CPU depth charts. But it comes into play I believe.