 |
View Old Forum Thread
Old Forum Index » Football Beta Testing » Football Beta Test Discussion » Suggestions on things to address before restart
lancereisen
Chris, the are several things that might be addressed before we do a restart.
Help me out here, guys.
Suggestions on things to address before restart
October 01, 2011 at 12:12AM View BBCode
If red, and the other big weeners [Chi, Win, NY] can sacrifice their juggernauts, so can Ariz... with trepidation. I've drafted enough baseball teams to know luck has a lot to do with it.:DChris, the are several things that might be addressed before we do a restart.
Help me out here, guys.
Admin
a) Changes to the initial population that provide for stronger defense and no longer supplies the 16 homogenized "starter" QB's. Also, the draft pool is deeper so it should have more overall talent on both sides of the ball.
b) Before the games get going I plan to put in some "self correcting" controls to keep one side of the ball from becoming dominant over the other side of the ball.
c) Improvements have been altered; we've seen the effects of that in Gamma this season but we now get to see it with the draft pool they were designed for.
I'm happy to take suggestions for other things that can be implemented in the couple of days before we roll.
--Chris
October 01, 2011 at 12:23AM View BBCode
What's being addressed here is:a) Changes to the initial population that provide for stronger defense and no longer supplies the 16 homogenized "starter" QB's. Also, the draft pool is deeper so it should have more overall talent on both sides of the ball.
b) Before the games get going I plan to put in some "self correcting" controls to keep one side of the ball from becoming dominant over the other side of the ball.
c) Improvements have been altered; we've seen the effects of that in Gamma this season but we now get to see it with the draft pool they were designed for.
I'm happy to take suggestions for other things that can be implemented in the couple of days before we roll.
--Chris
redcped
Just split up the first two rounds for the common positions, and then subsequent rounds just draw BA from the common pool of players.
Arguably, the first 3 rounds of LB should also be split into 2 OLB and 1 ILB ... although the future existence of the 3-4 probably means you'd need to make the 4th round also ILB, then the 5th round of LB would be BA.
October 01, 2011 at 02:46AM View BBCode
I would propose separate draft rounds for FS/SS and LT/RT.Just split up the first two rounds for the common positions, and then subsequent rounds just draw BA from the common pool of players.
Arguably, the first 3 rounds of LB should also be split into 2 OLB and 1 ILB ... although the future existence of the 3-4 probably means you'd need to make the 4th round also ILB, then the 5th round of LB would be BA.
redcped
The QB pool seems pretty deep to me. The top 35 guys are all making 4K or more, so everyone's going to have a very expensive backup if they want one.
And the top 25 QBs all range from 70-78 overall. To me that is very homogenized. Everyone is going to have a low A- or high B+ starting QB. With a backup nearly as good.
Plus, only 2 of the top 50 are under the age of 26. So everyone starts with a similar older QB save two teams that push the kids up the chart just to have a QB who is developing. And even if you go deeper in the group, there aren't many who are young enough to get much better.
Maybe I missed the point here, but what is this going to achieve? Shouldn't there be a few more young guys who need some development and a couple older A-level guys?
Seems there should be a real choice to start that is something like this:
1. I go after a stud who can win now but is already pushing 30.
2. I draft a guy who's 22-23 and develop him into a star.
3. I either don't get any of those guys because I bid low, or I decide to go with a solid guy who isn't quite as expensive but can be league average.
October 01, 2011 at 03:00AM View BBCode
Originally posted by Admin
What's being addressed here is:
a) Changes to the initial population that provide for stronger defense and no longer supplies the 16 homogenized "starter" QB's. Also, the draft pool is deeper so it should have more overall talent on both sides of the ball.
--Chris
The QB pool seems pretty deep to me. The top 35 guys are all making 4K or more, so everyone's going to have a very expensive backup if they want one.
And the top 25 QBs all range from 70-78 overall. To me that is very homogenized. Everyone is going to have a low A- or high B+ starting QB. With a backup nearly as good.
Plus, only 2 of the top 50 are under the age of 26. So everyone starts with a similar older QB save two teams that push the kids up the chart just to have a QB who is developing. And even if you go deeper in the group, there aren't many who are young enough to get much better.
Maybe I missed the point here, but what is this going to achieve? Shouldn't there be a few more young guys who need some development and a couple older A-level guys?
Seems there should be a real choice to start that is something like this:
1. I go after a stud who can win now but is already pushing 30.
2. I draft a guy who's 22-23 and develop him into a star.
3. I either don't get any of those guys because I bid low, or I decide to go with a solid guy who isn't quite as expensive but can be league average.
KLKRTR
October 01, 2011 at 07:45AM View BBCode
I'm very interested to see how the self-correcting tools work--especially after noting just how deep the WR and CB pools are..
casperthegm
Since there was concern that the passing game was more of an effect of wr vs cb or safety, does just addressing the qb talent level have the desired effect?
If wr and cb's are still equal in talent I worry it will still be a concern, unless this is addressed by the "auto correct". Personally, I would just like to have the coding for wr's changed so they just aren't quite as effective instead of turning to an artificial auto correct. Guess we'll see how it plays out...
October 01, 2011 at 01:12PM View BBCode
I'm generally in agreement regarding the qb's, especially the ages. I think that post by red is dead on.Since there was concern that the passing game was more of an effect of wr vs cb or safety, does just addressing the qb talent level have the desired effect?
If wr and cb's are still equal in talent I worry it will still be a concern, unless this is addressed by the "auto correct". Personally, I would just like to have the coding for wr's changed so they just aren't quite as effective instead of turning to an artificial auto correct. Guess we'll see how it plays out...
Fulla
October 01, 2011 at 02:43PM View BBCode
I just got started looking over the player pool. I see a potential problem with the QB pool where their ages are bunched up. Within the next 5 years, nearly every team will need to draft a young QB. This might not be an issue if the colleges were loaded with nice young QBs.
KLKRTR
October 01, 2011 at 05:13PM View BBCode
I think we'll wait and see what happens with the young QBs, but don't forget that any ages we're seeing here still aren't THAT old..I do agree that the WRs could be powered down a bit..
Admin
It's interesting that the pool looks homogenized without the "starter" QB's in the mix. I think that was just random chance, but I'll take a peek at the code. The intent is for there to be different "types" of QB's: some that are very accurate but low power, some good at the long bomb but less accurate, some that may not be great passers but are slippery and hard to sack. These might all end up with similar overall grades (and salaries) but still be very different QB's.
The purpose of a separate round for FB as opposed to RB is that a very different type of player makes a good FB as opposed to a regular RB. There's not as much of a differentiation in the tackle or LB positions. Safety there may be enough difference to make it worthwhile.
--Chris
October 03, 2011 at 05:16PM View BBCode
In the current code, initial QB's and K's are older on average than other players. RB's are younger on average (but that has always been the case). Part of the intent is indeed for the initial QB's to age out of the system faster.It's interesting that the pool looks homogenized without the "starter" QB's in the mix. I think that was just random chance, but I'll take a peek at the code. The intent is for there to be different "types" of QB's: some that are very accurate but low power, some good at the long bomb but less accurate, some that may not be great passers but are slippery and hard to sack. These might all end up with similar overall grades (and salaries) but still be very different QB's.
The purpose of a separate round for FB as opposed to RB is that a very different type of player makes a good FB as opposed to a regular RB. There's not as much of a differentiation in the tackle or LB positions. Safety there may be enough difference to make it worthwhile.
--Chris
lancereisen
I have bad dreams of a whole league of .500 teams, the teams with the most ties going to the SB.
October 04, 2011 at 03:30AM View BBCode
Can we elaborate on the 'limiting mechanism'?I have bad dreams of a whole league of .500 teams, the teams with the most ties going to the SB.
Admin
That would happen if the mechanism was applied game by game, but it will be applied week by week so the league averages stay within tolerances. Individual game averages will still vary widely. The effect should be pretty invisible, actually, if it is deployed at the beginning of a league. It would be dramatic if applied mid-season to an existing league.
There are a number of calculations for every element of the passing game, but each calculation starts with a "base number". If the league as a whole, for example, had an average QB rating 15 points above the NFL average, it may reduce the base completion percentage by 5 points. After all games are run, if the QB rating average has gone up, that correction factor would be higher for the next game, and vice versa, until equilibrium occurs. But the same factor is applied to all games and is linear in nature, so teams relative strength to each other is unaffected. This is what prevents the situation from deteriorating into a bunch of .500 teams.
--Chris
October 04, 2011 at 08:41AM View BBCode
Originally posted by lancereisen
Can we elaborate on the 'limiting mechanism'?
I have bad dreams of a whole league of .500 teams, the teams with the most ties going to the SB.
That would happen if the mechanism was applied game by game, but it will be applied week by week so the league averages stay within tolerances. Individual game averages will still vary widely. The effect should be pretty invisible, actually, if it is deployed at the beginning of a league. It would be dramatic if applied mid-season to an existing league.
There are a number of calculations for every element of the passing game, but each calculation starts with a "base number". If the league as a whole, for example, had an average QB rating 15 points above the NFL average, it may reduce the base completion percentage by 5 points. After all games are run, if the QB rating average has gone up, that correction factor would be higher for the next game, and vice versa, until equilibrium occurs. But the same factor is applied to all games and is linear in nature, so teams relative strength to each other is unaffected. This is what prevents the situation from deteriorating into a bunch of .500 teams.
--Chris
Fulla
If FB draft were moved to 3rd round, then round 2 is where everyone would grab the top FBs from the HB pool.
October 04, 2011 at 11:49AM View BBCode
Originally posted by lancereisen
I'd like to see FB drafted in the 3rd round.
If FB draft were moved to 3rd round, then round 2 is where everyone would grab the top FBs from the HB pool.
blakjakshalak
Having said that, this is a simulator after all. It's just numbers interacting. Sports is what it is because coaches and players are human and no computer code can replecate the spontanious interactions of humans making decisions (and mistakes) on the fly. So, perhaps this is the best solution.
October 04, 2011 at 12:23PM View BBCode
With all due respect, we have to ask ourselves if parity is the goal. Personally, I believe the goal should be an equality of opportunity, not an equality of outcomes. We can tweek the player generator all we like to address things like over inflated stats and the like. We all draw from the same player pool. But ideally, the outcomes of games and seasons should be, as much as is possible, up to the success or failure of the individual owners strategies.Having said that, this is a simulator after all. It's just numbers interacting. Sports is what it is because coaches and players are human and no computer code can replecate the spontanious interactions of humans making decisions (and mistakes) on the fly. So, perhaps this is the best solution.
casperthegm
If the coding works as intended, then the matchups should dictate the scores and outcomes rather than have an artificial auto-correct to keep things "in line" with expected results. Easy for me to say, since I don't have any insight on coding and how complex it is to get the matchups set properly, but in my humble opinion, that's what it should come down to.
Of course, it the mechanism is nearly as invisible as you say Chris, then I'm cool with giving it a shot and see how it works before passing final judgement. That's why we've got this league.
October 04, 2011 at 12:48PM View BBCode
I'm generally in agreement with blakjak. To me, it should be up to the players we draft, trade, and start on the field.If the coding works as intended, then the matchups should dictate the scores and outcomes rather than have an artificial auto-correct to keep things "in line" with expected results. Easy for me to say, since I don't have any insight on coding and how complex it is to get the matchups set properly, but in my humble opinion, that's what it should come down to.
Of course, it the mechanism is nearly as invisible as you say Chris, then I'm cool with giving it a shot and see how it works before passing final judgement. That's why we've got this league.
Admin
The matchups will still dictate results, there will just be a "scaling" effect to keep things from getting out of hand. Since it has low "granularity", the relative strengths between teams are preserved.
--Chris
October 05, 2011 at 04:13AM View BBCode
Originally posted by casperthegm
If the coding works as intended, then the matchups should dictate the scores and outcomes rather than have an artificial auto-correct to keep things "in line" with expected results.
The matchups will still dictate results, there will just be a "scaling" effect to keep things from getting out of hand. Since it has low "granularity", the relative strengths between teams are preserved.
--Chris
Pages: 1