Poisson Football Betting on Overs/Unders for Premier League: Week 2

We’Ve been looking at how well the passo model, you know where, if you look at the distribution of goals scored in football, they follow a Poisson distribution. So all we need to know is the average number of goals that our team scores and concedes, and then we can kind of predict. We can say how often we expect a team to score one, two, three, four or whatever goals. So using that approach we can look at a match, so in this case we’re looking at Watford Palace Arsenal, West Ham, Man, City Swansea, and then we can get the average number of goals that they score and concede from soccer stats.

The link is in the description below the video, so then we just plug that in to our pawsome model. We see how often we expect three or more goals or two or more girls Tector, because so far I’d always be looking at pretty clean the results. So you know whether it’s gon na be a home way in a drawer or away win.

This is the first look really to see. Ok, the passo model also does predict. You know how often we expect three or more goals etc, but is it accurate? I mean how do we use it here? I’M just gon na pick.

I’M gon na decide that okay, I’m gon na bet that we can have to feel more gold if the signal is 60 % or higher. On top of that, I’m using the history sucker starts to look at the historical rate of over and unders, because, okay, we can, we can get a prediction about Man City at home. You know cuz, they score lots of goals, but the other way is to totally ignore.

We don’t have any model, don’t don’t make it more complicated than you need to. So if you just look at the overs and unders, I can say, click at home Arsenal at home have three or more goals, 76 percent of the time. So I typically want to back three or more goals. If it’s Man, City at home, is Man City, eighty-eight percent of the time, so you know we could just use this and what the other game Iwas talking about was in Watford .., Watford, sixty five percent, so all three games, all three games seem to be high-scoring. Where we’d expect three or more, it is slightly more complicated, though, because Crystal Palace have three or more goals like when Crystal Palace are away less than half the time. So what’s the rules am I mean if the home team has to do of more goals?

Quite regularly, but the away team has a good defense and they tend to have low scoring game. I don’t know. Maybe I’d have to come up with the rule. We’D have to figure out what that means in this case, since I haven’t come up with a proper analysis, if anyone knows to be handy, what I’m doing is: oh because I keep man.

City are playing Swansea way and if you look at Swansea, they only have when Swansea are way they have low scoring games. They only have three or more goals. Thirty, five percent of the time.

So, even though Man City tend to score low circles, they’re playing against a team that tend to basically I should’ve shop, octopus or whatever you want to call it, so they tend to be involved in low scoring games, which means it’s not obvious that this is gon Na be a high scoring game. So that’s what we got from the overs and undress it’s on soccer strategy, just click on over San Andres and then we can get the goal rates from this table. So as we saw Arsenal at home, they score.

2 point 6. 5 concede one point: one they’re playing West Ham, who went there away score? One point: two four concede two point, two four, so getting those numbers we can plug them in to our pawsome model in our so for the Man City Swansea match.

We have two or more girls, 88 % of the time three or more 72. Now from just the overs and unders data, you know, if I didn’t have the model, you know I was saying: oh man, City score, 80, half three or more goals. 88 percent of the time, but Swansea away, half three or more girls, only 35 percent of the time. So if I was just using the overs and unders table, I wouldn’t I’d avoid the bedding on three or more goals. I mean it was unclear as to whether we should expect three or more. We could see that he both had two or more goals very regularly, but three or more.

If I just had the data, if I just had the overs and unders data, I would have not bet on them. I would have just bet on the two or more, but using the poisson model it says no, I know that Swansea have a good defense. You know – and they don’t tend to concede – that many away goals, but I think you’re gon na get fewer more very regularly, so the porter model says bet on man city swansea to have three or more. It also says to back arsenal West ham to have three or more eighty-seven percent of the time, so that just leaves Watford Crystal Palace. We have a difference.

So if you plug in the data for Watford Palace, I can see that we expect two or more girls, but it’s not clear about three or more. So I back two or more goals, but I’d stay away from three or more because it’s a coin toss I mean lucky, nearly fifty percent. So it’s it’s very much a coming to us it’s interesting because for Watford Palace we had a difference of opinion. We didn’t really know whether to buy three or more goals or not Watford at home, I think had typically had three or more goals, but Palace away only had three or more half the time.

So if I just had the soccer stats data, if I didn’t have the parcel model again again, I really wouldn’t know how to interpret that so very much like the Man City Swansea match where man said your home nearly always score. I mean nearly always have three or more goals. I think it was 88 %.

What Palace I mean, while Swansea way, rarely have three or more goals. So again, if I just had the soccerstats data, I would give that a miss, but using the poisson model it said. No, I think you’re gon na get three or more goals you nearly it had a very high probability of three or more so again, we’ll see how it goes, but today we have 10 euro on all three matches and we’ll compare the two approaches.

Just using the soccer stats information, if I just knew that old, this is how often the team has three or more goals, and I didn’t have the possible model, and if I had used the poisson model, does it actually add value?