ctriopelle
Footballguy
I'd guess that once the flyer score issue is found, you go up a few points. That is driving the top cutoff score up some
Last edited by a moderator:
Fantasy Football - Footballguys Forums
This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!
2008 $35,000 Subscriber Contest (3 Viewers)
- Thread starter Makavity
- Start date
Jayrod
Footballguy
17 points out of Dawson and you are probably still alive. 16 and there is still a chance (which is what he had last week). Plus if Buffalo gets a return TD, they could break 10 pts and help.The cut is likely to be around 161, so I wouldn't cash out just yet.
Jayrod
Footballguy
From what I can tell, the simulator probably produced a Bills Defense score somewhere in the 90's. That is making the crazy high top scores. If there are a few more defensive scores that are crazy high, it could really skew the results.
apalmer
Footballguy
Only if it's the Cleveland D. All the other scores are based on what actually occurred in the completed games.Anyone out there have the Cleveland D who can tell us if their score looks skewed too?
Last edited by a moderator:
Jayrod
Footballguy
Right, Cleveland or Buffalo. By others, I meant other runs of the simulator (he does it 1000 times)The one I identified was Buffalo....don't know if too many teams have Cleveland's D. (after searching, there are 20 with Cleveland's D alive and 6 that have both BUF & CLE and are still alive)
Last edited by a moderator:
nashua55
Footballguy
I like your chances better than mine. I am at 149 with Lindell(-6). I am on life support now. I need 4 forty yd field goals to have a chance. dont like my chances.
The Duff Man
Footballguy
Whew. Got knocked out in this week last season, so I am grateful to be moving on this year.
ditka...mike ditka
Footballguy
So I am confused....im at 153 with no one left...should I assume im eliminated or is there still an error giving extra points to some defenses?
ctriopelle
Footballguy
If you're spent at 153, yes, you are done, since the cut is already at 159ish. The error being discussed is in the projections, and it appears that for at least some of the simulations the Buffalo D is getting a score way above what anyone would consider as reasonable. The final cut per the sim for 80% of the runs will be between 159 and 162, so the cut will most likely end up in that range. If you can't get to that number by any method, you're toast (and you will show a 0% chance of advancing in the sim).
I have 169, so I think I will go through, but if I don't that will be why. That plus the A Bryant TD that got called back.
CrossEyed
Footballguy
That is a very real possibility for me. I'm at 159.45. Would be at 165.45 if the Troy recovery and TD would have counted.I don't mind not making it to the finals again, but it really stinks to have it be because the refs keep blowing calls.
Last edited by a moderator:
apalmer
Footballguy
They don't. That's posted by Doug in this thread (look back a page or two for the link).
jdoggydogg
Footballguy
:(And as porky Pig says......that's all folks! Doug's redo has me at 10% (was at 13%). I'll need roughly 25 out of KWII and 12 out of Buf D. Not very confident in that.The simulation estimate actually has required me to knock down it's score most weeks to hit the actual number (about 5% on average). This week, the pre-MNF cut is pretty much the same as the full value of the sample, probably because the remaining teams are getting more similar, therefore making the sample more accurate. Sorry to have gotten so many people's hopes up so high this week, including my own.
ctriopelle
Footballguy
I'd have made the last 3 FBG survivor style contest finals had the ref's not blown a call on Todd Heap in the last elimination week in the first version of the contest. The original format had tourney brackets for finals, and I can't remember how I would have fared at that point. Looks like this will be my earliest bow since we started doing this.
ahhh... got it, thanks... nice to be 100%... onto next week
wraith5
Footballguy
ginn coming on at the right time
Dirty Weasel
Footballguy
I think you may have miscounted. I have you down with a 10pt cushion. Also, the ownership percentages for BUF/CLE players are pretty low. You happen to own the BUF D, who are owned by 33.1%. They could raise the cutoff a decent amount all by themselves with a monster game, but you'd get the points. The next highest owned player is Lindell, at 12.8%. No other player is owned by more than 8.6%, so that's why your survival expectation is 100%.I'm 666 in the simwith only 7pt cushion
One of those things that I would have been better off not knowing about. Sheeeeeesh!
The Jerk
Footballguy
I'm done, too. Second year in a row I've gone out this week. 2006 I made the final cut but missed out on any winnings by a couple of points. Still a lot of fun. My only true disappointment is that I made every cut by nearly 10% over the level only to have the ENTIRE roster combined score 3 TDs this week. El Stinko.Alberta Clipper said:Fellow toast.... 151 Buf DT (-10) and Dawson (-6)
a_troll00 said:
Scarpaggio
Footballguy
Jerwol said:
189 and a portis/colston owner here.![[icon]](/data/avatars/m/5/5167.jpg?1660313464){kind=avatar}
Sheriff66
Footballguy
Colston owner here, and if I get booted, it will be because of him, total waste of $$$$.Jerwol said:189 and a portis/colston owner here.
Back to back 200+ weeks, good time to catch a streak!I think this team is strong......Tony Romo $23 20.90 32.10 17.90 32.70 26.50 34.05 0.00 0.00 0.00 0.00 Kurt Warner $5 15.45 36.05 20.60 32.60 25.50 22.70 0.00 30.05 28.90 34.40 Matt Ryan $4 13.85 6.90 17.10 9.00 22.30 21.15 0.00 24.25 23.10 24.20 Marion Barber $40 23.60 25.40 23.20 4.70 10.20 28.80 17.80 13.00 4.70 0.00 Thomas Jones $21 17.80 8.50 6.40 7.90 0.00 27.30 16.40 12.50 19.70 32.90 Ricky Williams $8 5.90 3.40 12.90 0.00 4.40 12.00 1.60 12.90 4.40 19.60 Ray Rice $4 9.80 0.00 2.10 0.00 0.00 12.20 6.90 11.60 19.10 2.90 Steve Slaton $1 6.40 0.00 20.40 21.60 22.10 6.60 16.40 13.60 15.80 3.40 T.J. Houshmandzadeh $34 7.40 5.60 32.60 11.00 27.50 11.90 13.80 14.30 13.50 0.00 Marques Colston $30 5.60 0.00 0.00 0.00 0.00 0.00 0.00 7.60 0.00 21.00 Brandon Marshall $23 0.00 40.60 27.50 19.60 5.50 18.80 14.40 0.00 4.70 20.90 Justin Gage $8 4.50 16.90 0.00 14.20 0.00 0.00 0.00 1.60 0.00 14.70 DeSean Jackson $4 16.60 17.00 9.10 21.60 3.10 15.90 0.00 10.20 4.30 18.50 Antonio Bryant $2 7.30 0.00 23.80 7.90 12.80 2.30 23.50 10.50 25.50 0.00 Tony Gonzalez $23 14.50 17.40 9.50 15.20 6.20 0.00 18.70 22.90 16.70 38.30 Jeff King $5 1.90 5.90 6.70 0.00 4.70 1.80 2.10 8.60 0.00 2.40 Dustin Keller $2 0.00 3.40 16.10 9.90 0.00 0.00 2.10 9.80 4.90 25.70 Nick Folk $5 4.00 16.00 10.00 7.00 8.00 9.00 2.00 10.00 2.00 0.00 Mike Nugent $1 2.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Chicago Bears $5 12.00 11.00 8.00 7.00 14.00 2.00 24.00 0.00 8.00 3.00 Miami Dolphins $ 1 5.00 2.00 8.00 0.00 2.00 9.00 5.00 12.00 13.00 3.00 Cincinnati Bengals $1 10.00 3.00 0.00 5.00 5.00 14.00 0.00 1.00 5.00 0.00 ---------------------------------------------------------------------------------------------------TOTAL 133.90 194.75 195.10 157.80 137.50 173.75 147.00 149.45 158.90 214.30 CUTOFF 112.85 132.60 133.60 138.70 122.20 139.65 121.30 138.55 135.10 153.00
Phil Elliott
Footballguy
At 136 with Lynch (0.10) so I would need a night like he had Week 3 with about 25 points to squeek in this week.
Investigation complete. The deal is that the lognormal distribution seems to model player (and team D) scoring very well in general. But it does have some issues at the upper end. About once every 200 "weeks," it'll spit out a score over 50 for a given team D, sometimes, as was noted, up into triple figures.
So I just manually capped it at 50 points. Not an elegant solution, but it works.
So here is a new set of simulation results at a new URL so you can compare against the old one if you want.
So I just manually capped it at 50 points. Not an elegant solution, but it works.
So here is a new set of simulation results at a new URL so you can compare against the old one if you want.
gman74
Footballguy
EBF
Footballguy
These factors conspired to kill me:
- I didn't own Steve Slaton or DeAngelo Williams. Both guys went off this week, which probably had a big impact on the cut. Not rostering Slaton was a gross error given his very low cost.
- Huge scores from Kurt Warner negated the presence of Peyton Manning on my roster. I have Warner, but so does everyone else. When I get a huge game from Manning, it doesn't help me because all the Warner owners are getting huge games for $5.
- Almost all of my "steals" are commonly held players: Portis, Calvin, ABryant, DeSean, Keller. The non-generic portion of my roster just didn't do enough to separate my team from the pack.
- Reggie Bush's injury was untimely. If he had been healthy for this week, he might have scored more than Portis in my RB2 spot. A 10-15 point difference there likely would've put me over the top.
Not a bad showing overall. In a contest like this where only about 5 people out of 10000+ entries earn significant money, you're looking at some LONG odds. You need to put together a great team and get super lucky.
- I didn't own Steve Slaton or DeAngelo Williams. Both guys went off this week, which probably had a big impact on the cut. Not rostering Slaton was a gross error given his very low cost.
- Huge scores from Kurt Warner negated the presence of Peyton Manning on my roster. I have Warner, but so does everyone else. When I get a huge game from Manning, it doesn't help me because all the Warner owners are getting huge games for $5.
- Almost all of my "steals" are commonly held players: Portis, Calvin, ABryant, DeSean, Keller. The non-generic portion of my roster just didn't do enough to separate my team from the pack.
- Reggie Bush's injury was untimely. If he had been healthy for this week, he might have scored more than Portis in my RB2 spot. A 10-15 point difference there likely would've put me over the top.
Not a bad showing overall. In a contest like this where only about 5 people out of 10000+ entries earn significant money, you're looking at some LONG odds. You need to put together a great team and get super lucky.
Last edited by a moderator:
Frank Costanza's Lawyer
Footballguy
Actually, I think the current format is the original format.2004 - Best Ball/Survivor2005 - Brackets2006 - Best Ball/Survivor2007 - Best Ball/Survivor2008 - Best Ball/Survivor
Alberta Clipper
Footballguy
tommyGunZ
Footballguy
I especially agree with your bold comments. Seems that the key is grabbing a few of the great values (Slaton, ABryant, Warner) and then hitting on a few oddball selections that few people took.Got my fingers crossed that some of the less popular players I rostered have big final pushes: WDunn, Bulger, Ginn, Marvin, D'Angelo, and DMason.
BuckeyeArt
Footballguy
As we get fewer and fewer teams left, it's more likely that FS's sample won't be representative of the whole contest and it be harder to predict the adjustment needed each week.
ctriopelle
Footballguy
Correct, but the way it was structured the 1st 2 times was different. The 2004 version was fixed roster design, with stiffer early cuts, and finals that were 64 team tourney style if I recall. 2005 was coordinated with another website, and I really don't recall the format. I hated it though as it took significant weekly research. 2006 was rostered like 2004, but the finals were like they are now, and I think the weekly cuts were still stiffer for a longer period (really penalizing bye week problems). Last year is when we switched to flexible rosters and the scored flex position, and the format was exactly like it is this year.Well, 5 minutes into the game and still a big zero from KWII...not looking good here...
ctriopelle
Footballguy
gave the best explanation I could earlier today, search up the thread to see what I came up with.
ctriopelle
Footballguy
Actually a "random" sample of 30 is sufficient, and we have over 120 live entries right now. Due to the fact it wasn't a true random sample to begin with is what caused me to have to add an adjustment factor. The unadjusted pre-MNF cut was within a point (and within the expected sample error) of our full cut value. I suspect that since teams are getting more and more similar, our database is becoming more like a true random sample of the remaining teams. Since the prior week required a healthy adjustment factor to get it correct, and I just use the prior factor until I get a new one, this is what made the call so wrong. I did also forget to fix the cut % to this week's number from last week's, but that was only about 1.5 points.
HellToupee
Footballguy
same here
bengalboy
Footballguy
Bravo
CrossEyed
Footballguy
And Polamalu's 12-yard return with a fumble on the last play of the game was overruled -- wrongly, referee Scott Green admitted afterward -- because of an illegal forward pass by San Diego as it tried desperately to keep a play going with laterals."The rule was misinterpreted," Green said.
ctriopelle
Footballguy
Well boys and girls, I have to get up early tomorrow, so I won't be doing any cutoff updates tonight. Based on how things look, I'd doubt it pushes beyond 161, right in line with the sim prediction. Good luck, and see you all next year, my ship has sunk...
