GUESS by the numbers Topic

Bob,

I am of the opinion that your current approach is basically right on and I am NOT in favor of normalization.  It also occurs to me that for QBs, it is typical that, relative to other players, QBs take more time to develop.  That is what we are seeing here, I think.

As long as the ratings are good at comparing WITHIN the same position, then I vote to leave it alone.  There is no real reason to compare one position rating vs another position.

Yeah, I know its nice to have a highly rated class via the GUESS reports....but its really not a huge deal.....
4/30/2013 10:00 PM
I agree with Harriswb3.  I like it the way it is.
5/1/2013 9:36 AM
The only thing this affects is the mythical GUESS recruiting national championship. If you're gonna start handing out Amazon gift cards, by all means fix it. Otherwise, save your time for the big update.

Thanks again for all your efforts!

5/1/2013 3:47 PM
I was thinking more about ACCEPTING Amazon Gift cards .... ;)
5/1/2013 5:36 PM
I will vote for no normalization as well Bob.  Thanks for all that you do in this game.
5/1/2013 6:45 PM
Simply put, my opinion is yes, something should be done to adjust the ratings of quarterbacks to make their worth more equal to that of other positions. 

I enjoyed Hazlewood's original post in this thread, but his mistake was the use of the word "normalizing." For some reason people have chosen to latch onto that word and can't get past it. It would help, as far as the results of this poll goes, if those people could see that the GD engine that "normalized" the player rating system for games played between two teams in seperate divisions is something entirely different than the idea of boosting the value of quarterback's GUESS ratings to more accurately reflect a real life ranking system and to not "reward" with a higher ranking those coaches who didn't have to sign a QB in their class. 

As Hazlewood said, QB cores being lower than that of other positions is clearly the cause of the problem. Looking at DII, because that is what I'm most familiar with, if you sign a QB with 50 GI,  50 Tec, and 50 Str, then you have signed a stud QB who you can hope to win NCs with down the road. If you take that same 50, and apply it to a RBs Spd, Str, and Elu, then we're talking about a player who is DIII caliber at best. Knowing this, it seems clear that something should be done to adjust for it when releasing the final rankings.

As the ranking system currently works, (again looking at DII) if three different teams each have one scholarship to fill, and the first signs the #110 rated DL, and the second signs the #60 rated LB, while the 3rd team signs the #1 QB in the country, that 3rd team who signed the star QB finishes lower in the rankings than both of the other two teams. I would say that should definitely be adjusted.

I would just like to add that I in no way meant this as any sort of slight to the current GUESS reports, and I apologize if it came off that way. GDreports is a huge part of what makes this game as entertaining as it is, and I don't think I would play any longer if they weren't around. I've said it several times before, but WIS should really be paying Hazlewood for what he does and I would love for the GUESS reports to be incorporated into the WIS site. The WIS recruiting rankings, as well as their Pre and Post season All-American teams are an absolute joke and need to be completely blown up and started over from scratch. 
5/2/2013 6:48 AM
fake, I totally agree with your entire last paragraph...without Bob and yatzr doing what they do, I would probably stop playing this game really quick.  Both of those guys add so much to the game that WIS should be compensating them in some way.

This still all boils down to the overall GUESS rating for your recruiting class or or your overall team.  And I still don't think it is necessary.  Your point is correct....a 50/50/50 QB at DII is a stud.  Everyone knows it and it will remain that way.  Why does he need to be compared to a RB of similar caliber?  Comparing QBs to QBs should be all that is necessary.

Now, when it comes to the overall team ratings, if I recall correctly, Bob assumes a Pro Set formation (vs Trips, Shotgun, WB, or ND Box) for all teams, and then gives some extra weight to the assumed starters.....or something along those lines.  So, if we are going to "normalize" individual players to get everyone on a level playing field...don't we need to then do something about the overall ratings process?  An all Shotgun team may have no RBs, therefore gets "penalized" when the system assumes that you have at least TWO starting RBs....even though you may have a fleet of stud WRs...

I'm NOT advocating that Bob change the overall rating system...I'm just trying to point out that by tinkering with one aspect of the ratings to satisfy a perceived "need" that same logic pattern could be applied elsewhere.  I just don't think that it makes the system/process "better"...it just gives Bob more work to do.  LOL

BTW, I'm still not fully awake yet and only 1 cup of coffee...so hopefully that ramble makes some sense....Hopefully all this debate is helpful to Bob.
5/2/2013 8:30 AM
Posted by bhazlewood on 4/30/2013 5:00:00 PM (view original):

I would normalize like this....

For QB: 70.38 / 58.90 = 1.195.  Multiply current QB ratings by 1.195 to normalize to RBs
For RB: Ratings get no adjustment.
For WR: 70.38 / 68.92 = 1.021.  Multiply current WR ratings by 1.021 to normalize to WRs.
etc, etc, etc,

Add a couple of columns to your information (normalized rating and normalized rank).

 

5/2/2013 2:13 PM
I'm all for normalization so that each position has the same 'GUESS Reports' core that is roughly similar.  That way the classes with punters, kickers and qb's can be judged the same as ones with a load of LB's and RB's.
5/2/2013 2:54 PM

Bob already ranks players, why not just give points for position rankings and divide by players to get a team value for overall and recruits. People have already discounted the relative importance of what the values mean for different players, just use the rankings.

5/2/2013 3:43 PM
why not simply take the qb guess rating and divide it by the highest rating of qb in that class giving you a percentage from .01-1? if you did this at every position you'd have easily comparable scores across positions.

5/2/2013 4:48 PM (edited)
A slight boost to your recruit class GUESS score is not going to have any effect on the actual game.  suppose ppl just like to see their team up in the recruit class rankings to raise their ego/ self-esteem a little bit. I think I speak for all, Im here to win NC's not #1 recruit class rankings skewed because my top ranked QB isnt ranked the same as the #1 RB.
5/2/2013 7:32 PM
Sure, in the grand scheme of things, how we finish in the recruiting rankings of a fictional football game does not matter in the least bit. For that matter, neither does "winning NC's" or anything else you can accomplish within the confines of this game. That is a given. Now that we have that out of the way, perhaps we can get past this whole "I don't care about recruit rankings I just want to win championships" mentality. It serves no purpose in this discussion. This thread was started by BH in an effort to gain feedback from the WIS community on how to put out the best possible product on his website. To say that you don't care not only seems offensive and counter-productive, but in all likelihood is also a lie. I don't think you can claim to love the GUESS reports in one sentence and then talk about how you don't care about them at all in the next. Those are two contradictory views.
 
Now maybe we can get back to the actual discussion of why these reports are, in the context of this game, important and why this adjustment to the rating system should be made. For starters, and most importantly, the main purpose of the website is to rank who has the best team and who had the best recruiting class as accurately as possible. That is what they are there for. If that isn't being done here, and the man has clearly identified why that is, then why not give him the go ahead to fix the issue? I have yet to read a reasonable opinion as to why things should remain as is. Not all change is bad, especially when it betters a product and nothing negative can come from it. Signing the best QB in the country should not negatively impact the ranking of your recruiting class. We all know that. 
 
Another reason why gdreports do matter is that they are a source of pride within this game. No one on here should be shamed for checking to see where their class finished in the rankings and then actually caring what they see when they do check. Their importance is all relevant, and this "I'm better than you because I don't care about the rankings" attitude is ridiculous. I can guarantee that some people get just as much joy out of a high GUESS ranking as they do from any CC appearance or big non-conference win. It's all part of the same game. I know that the first time I finished #1 in the GUESS recruiting rankings brought me as much pride as almost any single win I had had in GD up to that point. I would be willing to bet that just about anyone who has been around a while and secures their best ever finish in the rankings, or anyone who lands a top 10 class in their first year of a rebuild, feels the same way. A high spot in the rankings is like a reward for winning some of those crucial recruiting battles which can be far more intense than any actual game in WIS. So I would really like it if we could just drop the whole "doesn't matter" topic. I can't understand why people feel the need to go down that road in the first place. 
5/2/2013 9:12 PM
I just thought of another problem with all the methods of "normalization" I tried (and/or those suggested in this thread):

If you normalize as fatmoose suggested above - a methodology I also tried, but had forgotten about - you're saying that the best QB is always as good as the best RB is always as good as the best WR and etc. This is not true. Additionally, because of the big discrepancy in STDDEV at some positions, you still get a very dissatisfying mix of "Overall Top 50/100" recruits.
5/3/2013 4:36 PM
Posted by gt_deuce on 5/3/2013 4:36:00 PM (view original):
I just thought of another problem with all the methods of "normalization" I tried (and/or those suggested in this thread):

If you normalize as fatmoose suggested above - a methodology I also tried, but had forgotten about - you're saying that the best QB is always as good as the best RB is always as good as the best WR and etc. This is not true. Additionally, because of the big discrepancy in STDDEV at some positions, you still get a very dissatisfying mix of "Overall Top 50/100" recruits.

I understand what you are saying, but in the way the GUESS ratings are set now is that the best QB can never be numerically better than the best RB or WR because WIS just doesn't give them high enough values. Perhaps in the scheme of the game - the best players out there to recruit are OL or DL (look at the NFL draft) - but in GUESS reports they will never be #1 either. Bob is trying to counter the differences in WIS ratings to make the GUESS ratings more of an indication of the best recruiting class - to make his game within the game more enjoyable to the coaches (Thanks for the effort Bob - I much appreciate it.)

I personally don't look at the player scores very closely. I look at the rankings (overall and in class) to see how they stack up. When version 3.0 rolls out, with all the various player roles, it will be harder to compare values across all positions. The best blocking TE in the WCAA may score lower than the best receiving TE. Do you penalize the team with the best blocking TE? I say give them both XX points for having the #1 and XX-1 for #2 and so on. The values for each season will stay the same, QB, P, K and blocking TE's won't have to be normalized and we can still enjoy Bob's comparisons for bragging rights.

5/3/2013 5:27 PM
◂ Prev 1234 Next ▸
GUESS by the numbers Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2024 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.