Poisson model: weekend recap
All matches are played, so it is time to see how the models did.
Because I have 2 models per competition, we need to control which one does it the best. Not only I do have 2 Poisson models per competition but there are also 2 methods to based the predictions on. On one side, we could go with the percentages based on the model, but on the other side we could also based the bets on the most likely score predicted by the model.
The models
For all the competition below, I did offer these poisson model to the SBC community. The models there will be available a few days before they will become avaible on Steemit.
You can join us here: https://discord.gg/9dQJbVD
Per model I will post how the predictions were on the following betting types: 1X2, BTTS, O/U and Correct score.
Bundesliga
Here the model based on the current year is a little bit better than the one based on the last 3 seasons.
Ligue 1
Well the models did very well for the Ligue 1. The model based on the last 3 season even managed to get 3 correct scores! Impressive.
Serie A
I could have predicted better by just trowing dice. Just way to many goal for the Seria A and the Poissoin model. Both models were far from accurate.
La Liga
Just the stats. To tired to come up with some explanation!
Premier League
The models for the Premier League were also better than expected. Also here 3 correct scores in the model based on the current season!
Let me know if you like these stats!
Would like something different? Or do you just don't care?
Keep the faith,
Peter
Please put the clubs that you are betting on.
I don’t bet on all of these. On Friday’s I do place my bets online.
The model for this week will be probably online on Thursday.
People could use the model to be inspired.
I'm very interested to see how you calculate the distribution and apply it to sport scores. I could ask a ton of questions now but I'll wait till you post it first :-)
:)
Sir @fullcoverbetting,
I think the Premier League result is much better. Keep sharing sir. best of luck.
You can add the odds for the picks made by the model. The win % doesn't tell the whole story.
One of your models can make 60% correct predictions and be unprofitable, while the other can have just the 30% prediction accuracy and still make a profit. Cheers!
Something like this?
This one is for the Ligue 1
That looks good.
And if I'm not reading it wrong, the overall model did quite well this week :-)
It did for ligue 1. Other Leagues is a complete different story!
I expect that @djoker will have a theory as to why :-D
I don't have any point of view about that yet ^^.
However I agree on @beat-the-bookies point, what would be really interesting would be to compare the percentages given by your model and the odds. Then you can place a bet each time the odd is higher than the prob, and see if the model would be a long term win.
If you want to, I would also be interested in the way the calculations are done. (I could fully understand you do not want to share it yet)
I will give you some insights in our discord channel
You are correct, but that would quite a hassle!
Lets see what I can do about it. Will be something for next week! Thanks for the suggestion!
Congratulations @fullcoverbetting! You have completed some achievement on Steemit and have been rewarded with new badge(s) :
Click on any badge to view your own Board of Honor on SteemitBoard.
To support your work, I also upvoted your post!
For more information about SteemitBoard, click here
If you no longer want to receive notifications, reply to this comment with the word
STOP