Building a power rating system is a way a lot of people try to bet on sports. There are a lot of ways to go about doing this but most importantly you have to learn how to adjust your stats. Adjusting your stats helps take into consideration the strength of the opponent compared to the average team, and adjust stats upwards if your opponent is stronger than the league average, and downwards if the opponent is weaker. If done on a game-by-game basis for all teams in the league, you can create a system that generates season statistical averages that more accurately reflect the actual strengths of a team than if you were to run solely with raw, unadjusted stats.
This video is intentionally low level to help illustrate the absolute basics of adjusting stats at the most elementary level. This represents a fraction of a fraction of what you can do with adjusted stats. In this video I aim to introduce the basic core concepts of adjusting stats instead of an all encompassing tutorial.
Remember this channel is a tout free safe space. No tolerance for touts or speaking positively about touts.
At 934 if your taking the opponent Virginia’s stats and subtracting them wouldn’t the be an efficiency specifically against them and not the whole league? You said that’s be what they’re expected to score against the average team. Why would taking an opponents stats matter? That seems like it’s for that game specifically. Where am I off?
Possible to create a model for soccer correct score line ?
I’m having a hard time picturing how the stats would stabilize after a certain number of iterations. Wouldn’t Virginia’s offense just keep improving since FSU’s defense will continue to be better than average? Do the league averages change for each wave of adjustments?
i have a simple model that i teach. made 32 units last month and can be used in all basketball markets. hit me up if interested. mk1lima@gmail. c o m
What I don't understand how to turn stats into percentage of winning. For example if player is injured how do you calculate that into odds. Or if team is better by power ratings how do I know exact percentage of that team winning?
Thoughts on parlay cards?
I'm trying to figure out how you got possessions with those numbers and I can't get them to add up. Any insight on how you calculated it? Thanks. Great video.
Would it still be beneficial to include last years stats in a newly made model to predict current year? How does graduation/draft incorporate into a college BBall model?
How would you calculate H3 for football?
Would you say that adjusting a team's raw stats game by game in order of the season schedule is more effective than simply taking the raw stats and adjusting them based on their average opponent played for the year? So in this example, adjusting each team's OE and DE game by game (based on that particular opponent's DE/OE) VS adjusting each team's raw OE and DE by average opponent's DE/OE differential (for that season). Thanks!
Do you have to use OE and DE as the stat you're adjusting, or can this be done with any statistic?
Just found your channel recently, and have been very impressed with your approach to this activity. I was wondering though if you could go more into adjustment types? I was wondering what the difference was with your subtraction method adjustment versus what I've seen in other articles where they take the "(team off value / avg def value) * opponent def value)". Thanks and keep up the great work!
Is that correct? In your first video of the series, (intro to expectation), you added the calculated deltas to the league average. But in this video, you are subtracting the deltas? How would I go about adjusting a team's season stats?
Why would you use last years stats to create a model for college hoops when the players change every year?
This is just superb, I have been researching "all sports betting" for a while now, and I think this has helped. You ever tried – Ameliaten Bankify Bigshot – (just google it ) ? Ive heard some great things about it and my colleague got amazing success with it.
I have betting model is all i creat by match time and etc, comment me, if you like to speak to me accouracy 90 percentage draw full time coorect score etc
Can this be used to adjust points for and points against? Or should it only be used for other stats. Someone please help.
the question is: which model do the bookies use?
"EXPECIALLY"
Would this method be still relevant in a league with very few games, I want to make a power ranking of the Mexican football (soccer) league but there are only 18 teams and they only play each other once before playoffs.