Right. As you know, sometimes we here like to run Monte Carlo simulators to estimate likelihoods of things happening.

What?

So. I simulate 50,000 seasons game by game. The idea is that if, in any given game, Fulham have a 27% chance of winning, a 21% chance of drawing and a 52% chance of losing (as has been our form in the last 52 games), and if you then run 38 games like this, and repeat the episode over and over and over you can see how likely things are to happen, because in ‘only’ 38 games a wide range of things can happen.

The result?

In those simulations where the team had 10 points in 14 games, the most likely end of season outcome was…. 33 points. Â A likely range is maybe 31-37.

Oh.

40 points or more? That happens 23% of the time.

This assumes that the team is going to be at its level of the past year or so. Â That means a bit better than this season’s efforts but not as good as we might like. Â Now, Rene may very well improve the team but how much room for growth is there? Â We can want all we want but it seems optimistic to expect a performance level established over 52 games to suddenly be far too low, new coach or otherwise.

So we are in trouble. We knew that, but here it is statistically.

### Like this:

Like Loading...

*Related*

Fair enough, but 37 would quite probably be fine, 36 could well be, and even 35 might be, so still plenty of scope. Let’s see if the improvement of yesterday is sustained, because if it isn’t we are going to fail by miles, Monte Carlos not required, but it is we will be in with a shout, Monte Carlos nothwithstanding. Pre-match yesterday I assumed we were doomed as there seemed no prospect of winning every third match (on average). Now it just seems a tall order.

But – normally with 37 points you need goal difference on your side – which we do not have at the moment :-(

There are 2 methods that I measure us by. To stay up we need a point a game average (38 point is normally ok). And to get us around that 8-12 place rage we normally have the same amount of wins as draws as losses (so roughly 12 wins, 12 draws, 12 losses = 48 points).

So we have 24 games left:

– point a game: 24 + 10 (that we already have) = 34 points (about the same as monti carlo), relegated

– win/draw/loss ratio the same = 8 wins + 8 draws = 24 + 8 + 10 = 42 which is safe!

We have a chance, we just have to play the rest of the season like we do when we finish about 12th.

Ultimately (stating the bloomin obvious) we need three teams below us come May. I believe that is totally doable provided we play more like we did against Spurs and less like we did against West Ham. I am most optimistic.

Rich, could you run a comparison of this model vs where we were at this same point in the Great Escape, as wells as comparisons of when Roy took over (points earned, games remaining, prior form, etc.)? You know, for funsies- or is that too much?

Sure will try that today. I know we got to about a 5% chance of survival at one point

After Roy took over we gained 22 points from 18 games (average 1.222), but twelve of those points came in the last five matches.

As I see the present situation, Sunderland and Palace are now improving as, I hope, are we. Above us Cardiff, Norwich, West Ham, Hull and Stoke are all a bit iffy.

Norwich have have had the hardest fixtures in the league. We have had among the easiest.

Palace prob still doomed as financially are a long way behind. Sunderland have a huge hole. All three of us to go?

I think we’re getting to the point were the “hardest” fixture stats (ie Qualcomp) is about to be reset or made null considering it’s a balanced schedule. I need to flesh out this idea more, but Strength of Schedule really only “works” when teams haven’t played the same opponents, and the league is a month away from that equilibrium. For example, West Ham’s Qualcomp was like #2 and dropped to #11 after playing us.

Anywho, Cardiff and Hull are going to stop running on adrenaline and regress to the mean. Despite their results, their TSR is already really bad, and I don’t see it improving.

Conversely, I do see our improving if for the only reason the players aren’t playing to get their coach fired. This all feels like USC’s recent football season with Lane Kiffin.

I think Steve Bruce might be canny enough to pull off a survival and I like Huddlestone’s ability to run the midfield. Cardiff spent quite well and may also be okay. Going to be tough.

Simple fact is that we need to overtake one of the teams that is 3 or 4 points better off than us and stay above Palace and Sunderland. We play all of those teams at least once so the deficit can be overcome in one match.

Sod statistics. That’s very doable if we improve and impossible if we don’t.

Alright alright. Someone asked me how many points I expected and I had no idea. Now I do and I don’t think it will be enough, improvements or not.

But the more we improve the more points we will get and statistics can say nothing about the likelihood of improvement in a situation where a failing manager is replaced by someone who knows what he’s doing.

It’s whether you believe any changes can happen that quickly. All we are saying here is that with reversion to something like last year’s form we are still in trouble (“I don’t need statistics to tell me that….”).

I don’t know that we have it in us to suddenly get to a higher level than that. Yes in many ways this is pointless but so is any attempt at forecasting when variables are unknown. You either take it in or you don’t. I found this useful; others do; others don’t. All are of course fine.

Nothing wrong with the original concept, Rich, it’s just untimely, the key to our fate being whether the performance level and style we saw the other night is a one-off or not. So much depends on whether the past half season is or is not a guide to the next half season and we’ll soon enough know.

I’m optimistic simply because I believe it IS possible to get that 52% loss rate down, even if it’s more draws than wins.

COYW :)

Hiya! I simply wish to give a huge thumbs up for the great data you¡¦ve here on this post.

I might be coming again to your weblog for more soon.

Excellent confident synthetic attention for the purpose of detail and can anticipate difficulties before these people occur.