×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

75% Complete (with a PS to alta)

Shadow Wrought (586631) writes | more than 8 years ago

Moon 23

Well we are through three quarters of the season and, entering the fourth, I stand at 108-84 on the year, which is a .562 winning percentage. Even though I'm, or rather the dice, are 24 games over 500 at this point, with 48 games left, I am still far from assured a winning record. This is especially true because of the problem I mentioned earlier- namely that these results are based on how strong teams were atWell we are through three quarters of the season and, entering the fourth, I stand at 108-84 on the year, which is a .562 winning percentage. Even though I'm, or rather the dice, are 24 games over 500 at this point, with 48 games left, I am still far from assured a winning record. This is especially true because of the problem I mentioned earlier- namely that these results are based on how strong teams were at the start of the season. To put this in perspective I created the following chart which shows the die that each team rolled at the beginning of the year (either a d4, d6, d8, d10, d12 or d20) and the die that they would currently earn. (All of these are based off of Dr. Z's Power Rankings.)

Team -- Initial Die -- Current Die -- Change
Arizona Cardinals -- 6 -- 6 -- 0
Atlanta Falcons -- 12 -- 8 -- (4)
Baltimore Ravens -- 12 -- 4 -- (8)
Buffalo Bills -- 8 -- 6 -- (2)
Carolina Panthers -- 20 -- 12 -- (8)
Chicago Bears -- 4 -- 12 -- 8
Cincinnati Bengals -- 8 -- 12 -- 4
Cleveland Browns -- 4 -- 6 -- 2
Dallas Cowboys -- 10 -- 10 -- 0
Denver Broncos -- 8 -- 12 -- 4
Detroit Lions -- 8 -- 4 -- (4)
Green Bay Packers -- 8 -- 4 -- (4)
Houston Texans -- 6 -- 4 -- (2)
Indianapolis Colts -- 20 -- 20 -- 0
Jacksonville Jaguars -- 10 -- 12 -- 2
Kansas City Chiefs -- 10 -- 10 -- 0
Miami Dolphins -- 6 -- 8 -- 2
Minnesota Vikings -- 10 -- 10 -- 0
New England Patriots -- 12 -- 8 -- (4)
New Orleans Saints -- 6 -- 6 -- 0
New York Giants -- 4 -- 12 -- 8
New York Jets -- 12 -- 4 -- (8)
Oakland Raiders -- 6 -- 6 -- 0
Philadelphia Eagles -- 12 -- 8 -- (4)
Pittsburgh Steelers -- 12 -- 10 -- (2)
San Diego Chargers -- 10 -- 10 -- 0
San Francisco 49ers -- 4 -- 4 -- 0
Seattle Seahawks -- 8 -- 20 -- 12
St. Louis Rams -- 10 -- 8 -- (2)
Tampa Bay Bucs -- 6 -- 10 -- 4
Tennessee Titans -- 4 -- 6 -- 2
Washington Redskins -- 4 -- 8 -- 4

As you can see only 8 teams, an even quarter of the field, had no change. The rest either went up or down, sometimes dramtically (like the 'Hawks and the Ravens), or just a bit. So as the season draws nigh, the stronger teams will get stronger and the weaker teams will get weaker. If that has flip-flopped from the beginning of the season, then the built in advantage for the stronger teams has become a deficit. So trying to weight the dice in the stronger teams favor has actually made the result less likely if the teams reveresed. That means that luck will be playing a much larger part in this for final four games. When the final playoff teams are decided, I will play out the playoffs for you the same way, and we'll see how it works itself out. My Super Bowl predicion du jour? Colts win it (completing their perfect season) 14-6 over... (*drum roll*) Chicago. Sorry 'Hawks, but Chicago takes it from you in a heartbreaker. (Well, in this mornings vision they do. Stay tuned for tomorrows when you beat the Colts in OT;-)

Finally, I give you this weeks predicitons and my insightful comments ont he match-ups...

Matchup -- Predicted

Washington at Arizona -- Arizona (Why not? Arizona has years more experience at mediocrity, but Snyder is doing his best to get Washington there.)

New Orleans at Atlanta -- Atlanta (Atlanta needs all the wins they can get if they want to make the playoffs, it all hinges on whether Good Aaron Brooks shows up or Bad Arron Brooks shows up. There's just no telling.)

Tampa Bay at Carolina -- Carolina (This'll be a good game, full of playoff implications. My gut tells me that Tampa pulls out a close one.)

Cleveland at Cincinnati -- Cleveland (This reminds me of those great college rivalry games where one team is ranked high and the other isn't ranked at all. Everyone talks about how tough a game it will be and so forth, and then it either ends in a 19-17 upset or 66-3 thumping. This one'll be the latter.)

Kansas City at Dallas -- Dallas (Lot's of Larry Johnson and Drew Bledsoe running in the same direction.)

Baltimore at Denver -- Denver (Denver brings kindling to this game. Places it under Brian Bilicks chair afterwards.)

Detroit at Green Bay -- Detroit (There is no, repeat, no truth to the rumors that Ohio State will be filling in for the Lions. In other news, RetireWatch 2005 enters its 14th week...)

Indianapolis at Jacksonville -- Indianapolis (I soo want the dice to be right. How cool would that be! Jax finds out how tough Indy is to beat.)

New England at Buffalo -- New England (Losman has a breakout day while Brady throws 5 picks. Well maybe not, but I think Buffalo takes this one.)

Oakland at NY Jets -- NY Jets (Oakland had its first game without a penalty since 1971 in their Sunday drubbing against the Chargers. If they get flagged against the Jets, they win.)

NY Giants at Philadelphia -- Philadelphia (Their season over, their pride in little shattered pieces by their feet, Philly opens up a desperation can of whoop ass against the Giants. Luckily the Giants miss 7 field goals so Philly takes it 9-3.)

Chicago at Pittsburgh -- Pittsburgh (Chicago is a really, really bad team to play when you are trying to figure out your offense.)

Miami at San Diego -- San Diego (Even though Saban has won more this year than the 'Fins did last year, I just find myself extremely underwhelmed by him. I think he'll coach another four years and remain in the depths of mediocrity the entire time.)

San Francisco at Seattle -- Seattle (Ah yes, the famous "Trap Game." This is marketspeak for, "Yeah but if SF does win, you'll kick yourself for not watching.)

St. Louis at Minnesota -- St. Louis (Obviously the Rams failed to read this last week. The Juggernaught starts, NOW!! Humor aside, this is certainly a winnable game for the Rams- I'm not sold on the Vikes- but they have to actually decide to play.)

Houston at Tennessee -- Tennessee (Another "Trap Game." Eyes on the prize {Reggie Bush} Houston, eye on the prizes {boatload o' draftpicks}.)

PS: To alta, I friended you because of this comment in reply to my Xbox360 Poll post. It was exactly the way that I would have done it currently and, because I thought that that was cool, I friended you. I do these things off of gut feelings, and, after reading that, my gut said to pull the trigger.

23 comments

Hey. (1)

artifex2004 (766107) | more than 8 years ago | (#14203928)

You should try your hand at predicting sports at owise [owise.com] .
Free and you can win Amazon gift certificates.

Re:Hey. (1)

Shadow Wrought (586631) | more than 8 years ago | (#14204315)

Contributer, Level 5 [owise.com] ;-) I was wrong on a couple Monday Night "Locks." The most frsutrating of which was when Dallas gave up two touchdowns (and the game) in the last minute and a half or so. Grrrr. Overall I flit between the top of the Contributer ranks and the bottom of the Analyst ranks. I peaked at 4th place overall, then got cocky:-(

Now I just go every Thursday, look at what's going to expire the soonest, and just go with my gut. I've given up on the Amazon cards and am now just testing myself;-)

Re:Hey. (1)

artifex2004 (766107) | more than 8 years ago | (#14204882)

I forget who's joined, sorry :)
The first period, I was at or near #1 overall until the very last day of the contest, and then a bunch of movie results came in and killed me. You can now see me annoying drang with my stupid questions, sometimes. My overall weight is .06 now. Hey, at last something I'm underweight in. :) I've got a plan for recovering some rank but it's very very risky. Oh, and me know anything about military or sports? hah. Maybe that's why I do less poorly on them, though, because I research more.

I just realized... (1)

FortKnox (169099) | more than 8 years ago | (#14203966)

... you are doing this all by hand.

I should make a program outta this and even put in the AI behind it to change the dice around and give it to you so you have something to look at while you are learning how to program...

Re:I just realized... (1)

Shadow Wrought (586631) | more than 8 years ago | (#14204284)

After doing it I just realized that I could make Perl do it for me! If you feel inclined by all means throw it together, but it will be one of the thigns I try to create. I will still likely do a lot of it by hand because, well, I like rolling dice;-) It will also be interesting to compare how the program rolls its dice as compared to my rolls. I worry that one of the things I might try is to refine it too much. For example, Perl is not limited by common dice on the market. So instead of the leap between d12 to d20, I could instead simply have eight levels of "virtual dice," so you'd have d4, d6, d8, d10, d12, d14, d16, and d18 or the like. Only time will tell;-)

Re:I just realized... (1)

FortKnox (169099) | more than 8 years ago | (#14204390)

Throw in a neural network, and you could really be in for some seriously wierd dice (specialized in AI at school... its how I think).

Re:I just realized... (1)

Shadow Wrought (586631) | more than 8 years ago | (#14204425)

Very cool. I have thought about AI (I have 0 knowledge of it) and I thought through how I would try and develop one for my top-down tabletop clone of a video game. It was, and is, a fun thought process, so I will hopefully be able to pick your brain when I get closer to implementing it;-)

Re:I just realized... (1)

FortKnox (169099) | more than 8 years ago | (#14204708)

Well the AI I am talking about it neural networks and genetic algorithms and stuff. Not stuff you use in games (mostly). Game AI is done with agents... you teach it a variety of things to do, randomize it, add the ability to make it smart or dumb, and let it go.
Neural networks are for pattern recognition and genetic algorithms are basically the chaos theory of the CS world ;)

Re:I just realized... (1)

Shadow Wrought (586631) | more than 8 years ago | (#14204862)

To borrow a quote from Hot Shots!: "I don't have a clue what you're talkin' about, FK. Not a fucking clue;-) Though I wonder... It might be fun to make my own version of the BCS for the NFL;-)

Re:I just realized... (1)

NFLFan (878402) | more than 8 years ago | (#14211344)

I just read something the other day about how they predict the weather. Seems they use bunch of different models/simulations and they usually converge as the weather gets closer-- giving them their forecast.

I might be fun to do something like that with picking this games.

Re:I just realized... (1)

FortKnox (169099) | more than 8 years ago | (#14211628)

Given enough information, you can really do a lot.
First step is just the record.
Next you put in the 'major' stats, like yardage, score, etc..
Then you look at home and away games.
Then you look at individual players and how well they play in grass vs turf, open field vs dome, the weather, etc etc etc...

The more you put in, the more accurate the predictor will be.

Doing the weather model, you could have different neural nets focusing on various items. Like one would be concerned with the field, one would be concerned with the weather, one would be concerned with the record and major stats, and one could be concerned with individual players...
Then you have one to determine how much weight to put on each system....

But that's a hell of a thing to program...

I wonder (1)

sillypixie (696077) | more than 8 years ago | (#14203967)

Why alta would post a JE asking to know why people friended him, and then disable comments on that journal?

Doesn't that seem self-defeating?

Oh well.

Re:I wonder (1)

Shadow Wrought (586631) | more than 8 years ago | (#14204337)

Yeah I thought that was kinda weird, too, and I'm really not sure how he would like to be notified, but what are you going to do? Maybe I should have invited everyone who's friended alta recently to post their reasons here... ;-)

Re:I wonder (1)

sillypixie (696077) | more than 8 years ago | (#14204622)

No publicly shown email either. I guess enabling comments on a single JE isn't a perfect answer either, since it would eventually be archived.

Oh well! Enough of that dilemma, back to solving world hunger (-:

pix

Re:I wonder (1)

StalinsNotDead (764374) | more than 8 years ago | (#14204934)

Slashdot can notify you of any relationship changes. I get mine via email.

The email includes a link to that person's page on slashdot with the title and a couple lines of the person's last journal entry and their last few comments.

Re:I wonder (1)

Shadow Wrought (586631) | more than 8 years ago | (#14205097)

The weird thing was his JE, basically wondering why people had friended him- with comments disabled. If he had enabled the comments I, and likely anyone else who had added him, would have simply replied. So he was basically asking people to tell him why they had friended him without really giving them a means to do so;-)

Re:I wonder (1)

StalinsNotDead (764374) | more than 8 years ago | (#14205150)

It appears do be his only journal entry. Maybe the comment options drop down list defaults to comments disabled, and he forgot to enable them.

Rumors (1)

RailGunner (554645) | more than 8 years ago | (#14204268)

There is no, repeat, no truth to the rumors that Ohio State will be filling in for the Lions.

Oh sure, get MY hopes up.

Once all the bowl games are done I'll post my percentage. I did see that Ohio State is a touchdown favorite over Notre Dame, and that USC is favored over Texas... still sticking with my pick of Texas over USC.

Re:Rumors (1)

Shadow Wrought (586631) | more than 8 years ago | (#14204400)

Put that one in just for you;-) I'm surprised Ohio State is only a 7 point favorite. I think ND is way over-rated and I am hoping that a) Ohio State just routs them and b) that Oregon stomps the Sooners. I can live without a playoff, but can we please, pleeeeeeeeeeeeeaaaaaaaaase get rid of the automatic bid crap?

football outsiders (-1)

shitface (121619) | more than 8 years ago | (#14205327)

Check out http://www.footballoutsiders.com [footballoutsiders.com] . I admit to not really understanding their DVOA rating system but the ratings seem to always hold with my beliefs so it seems to just work. I have been toying around with the idea of creating some sort random equations for picking games with all those numbers. I think the outsiders would be a more objective than Dr. Z.

If strength of schedule is as important as some people seem to suggest it is then it could be interesting to use your initial findings as some sort of way to estimate schedule difficulty and use that to come with your own rankings and roll the dice on that. I would think such a system would give more consistent results for the whole season.

impressed (1)

NFLFan (878402) | more than 8 years ago | (#14211297)

I have been surprised how well your random picks have done.

Sort of reminds me of somebody trying to sell paintings done by infants as modern art for outrageous prices. How good is a guy who can pick winning teams at a 60% clip? Probably not as good as what we all think. Just order the teams from best to worst as you perceive them, or someone else like Dr. Z, and then roll some dice with the ordering as some sort of weight system and ta-da you can be right around that 60% clip, at least you are. The ordering might be the hard part but because you are using a monte carlo method the ordering does not have to be exact.

Re:impressed (1)

Shadow Wrought (586631) | more than 8 years ago | (#14211902)

I have been surprised how well your random picks have done.

Thanks! Frankly it has shocked the shit out of me how well it still working. Even though I don't think I'll be going much over .500 for the last four weeks, I was really expecting to be writing weekly columns of going 2-14 and such. What will be more interesting is next year seeing if I can add a line factor into it, and see how well it would do if I were actually betting all of this;-)

Check for New Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...