Three weeks in. I opened my Pick ‘em and I see Menard and Rocksprings at top of list. I pick Rocksprings and go about my business. Checking my guesses against The Toy, I see immediately that The Toy and I disagree on the first one. I expect some bad mouthing from SFA and a few more of my friends on here, especially from Menard, but I think the Coach Clutch led team from Rocksprings will win this one by 20. SFA, if you will pick me up we will go to this one. I wish The Wesp’s mom would take him.
Time to pick on the Toy etc.
- Thread starter oldfolks
- Start date
To be fair, that particular game has nothing to do with The Toy, and everything to do with how low I had Rocksprings to start the season. I didn't really know anything about their team so, pretty much like all the rest of the pre-season picks I do, I just guessed. 
Evidently I guessed that one way wrong, but they're making big moves up over the last few weeks.
Evidently I guessed that one way wrong, but they're making big moves up over the last few weeks.But we like to pick on The Toy.To be fair, that particular game has nothing to do with The Toy, and everything to do with how low I had Rocksprings to start the season. I didn't really know anything about their team so, pretty much like all the rest of the pre-season picks I do, I just guessed.
Well don't let me stop you. Carry on!
I mean, Bluff Dale beat the spread by 40 points, against a top 5 team, and then barely moved in the rankings.
Texlonghorn75
Six-man expert
Bluff Dale needs to meet and beat May to move up. Lol Beating the spread and losing the game proves very little but some grit.I mean, Bluff Dale beat the spread by 40 points, against a top 5 team, and then barely moved in the rankings.
sillyjeeper
Six-man observer
The toy picked Jonesboro to lose to Union Hill by 28 pts. Jonesboro 45'd them according to the score on here. The only time I look at the toy is for picks for all the private/TAAPs divisions pick 'ems. And I like to beat the toy in my pick ems, one last chance to fight and win against the computers before they take over the world.
Actually, The Toy is usually much better at picking UIL games than non-UIL games.
In 2023 The Toy guessed right on 81.97% of all UIL vs UIL games, 79.72% of non-UIL vs non-UIL games, and 78.97% of non-UIL vs non-UIL games.
Last year The Toy guessed right on 81.17% of UIL vs UIL games, 79.28% of all non-UIL vs non-UIL games, and 77.14% of all UIL vs non-UIL games.
Last week The Toy guessed right on 85.96% of UIL vs UIL games, 70% of non-UIL vs non-UIL games, and 73.49% of UIL vs non-UIL games.
Another interesting stat for those of you who are interested in this kind of thing. Last week The Toy picked games to within a margin of error of 40.87 points per game. That's how many points the predicted spread is away from the actual spread for each game. Now, before y'all start screaming about how terrible that is, consider this. If a team is picked to "win" by 100 points, and they "only" win by 50, the system still calculates that margin of error at 50. I'd originally planned on making the system so that if a team is picked to win by 45, and they do it, the error would always be 0, but I thought it was a lot more interesting to see how close it was to picking the exact spread, rather than just TeamA will 45 TeamB.
Of course it's not a perfect system, but that doesn't exist. It's still one of the best systems, if not the best system, in place for picking six-man football games. Also worth noting, because it was brought up, is the fact that the pick 'em games that are picked each week are 30 of the games that The Toy expects to be the toughest to pick a winner on. No game with a predicted spread over 40 is added to the pick 'em pool.
I’m not sure Jeff Sagarin himself could develop a really effective model for predicting 6-man outcomes. The mercy rule is difficult to account for. Did team A send team B home at the half or late in the 4th Qtr. Two different outcomes. Hard to differentiate teams based on common opponents. Elite and very good teams will 45 a common opponent, but it doesn’t tell you much. In most cases the elite team will turn right around and 45 the very good teams. For the average teams we see team A beat team B who beats team C who in turn beats team A. So many times the games hinge on individual matchups and that is hard to model. As you get in more games I would expect an iterative model to sharpen up a bit. Never the less, I would not try to make a living betting 6-man football games. Small schools, small rosters, and a game that can be greatly affected by one or two top athletes.
I mean...a lot more was proven on the field that day, but that's ok. I've moved on.
Tigerfan85
Six-man master
Losing to RS was all part of the master plan to stay out of the Toy’s sights. In case you were wondering why we flubbed that up.
sillyjeeper
Six-man observer
Then the toy goes and has a perfect pick on the Rankin vs Westbrook game picking Rankin by 4. 
Rankin won 48-44. Maybe that darn toy knows what it is talking about after all.
Rankin won 48-44. Maybe that darn toy knows what it is talking about after all.
Reddevil4life23
Six-man fan
What a gameThen the toy goes and has a perfect pick on the Rankin vs Westbrook game picking Rankin by 4.
Glad to know that so I can make a little tweak next week.
You know what they say about squirrels and nuts.Then the toy goes and has a perfect pick on the Rankin vs Westbrook game picking Rankin by 4.
Wildcats5421
Six-man expert
I agree dude it was a great game congrats to the Devils. We will be back but what a night for football
Tigerfan85
Six-man master
I see what ya did there this week with the Toy. You got jokes.
Wildcats5421
Six-man expert
lol he did something, how does Jonesboro leap a team they got 45’d by
There are always questions. It's never perfect. The best answer I can give is that The Toy is rating teams as they are "right now". Not that the past games don't play into it at all, because they definitely do, but Jonesboro 45ing Union Hill this week holds a lot more weight than the Westbrook/Jonesboro game last week. May beat Whiteface in week 1 and has been ranked below them since week 2. I'm sure there are many other examples of past games maybe not carrying as much weight as they should when determining the current ratings, but I believe that's by design. The Toy can't determine injuries, new players coming into the mix, etc, so just adjusting ratings by recency over everything else is really the best way to deal with that.lol he did something, how does Jonesboro leap a team they got 45’d by
wildcatfan4
Six-man pro
This is why I don’t follow the toy on rankings or spread. WB not only beat Jonesboro by 45 but they traveled 3 hours to beat them at their own place.lol he did something, how does Jonesboro leap a team they got 45’d by
I get it if the rankings down the road moved them above for whatever reason but this was 7 days ago. Haha
Also important to realize that The Toy doesn't know Westbrook beat Jonesboro. It knows that TeamA, rated X.xx, beat TeamB, rated Y.yy, and it adjusts based on that. It's all completely mathematical based on predicted spread vs actual spread, and it only calculates one game at a time. Union Hill was rated quite a bit higher than Jonesboro, then Jonesboro 45d them, so they got a huge rating increase based on that. The Toy doesn't know not to increase their rating so high that they jump a team they've already been beat by, even if that game was yesterday.
Obviously this is one of the downsides of having a rating system that can only take numbers into consideration.
