Points Required For Survival

All advertisments are hidden for logged in members, why not log in/register?


Silent Blade said:
Try Norwich to lose every match except against us, also Watford (or West Ham) to lose every match and it is more than 8 points
Click to expand...
Yep you can't just give those at the bottom maximum points and those already safe none, it doesn't work like that.
 
We aren't going down, close the thread and send the link to Norfolkinchance.
 
Silent Blade said:
Try Norwich to lose every match except against us, also Watford (or West Ham) to lose every match and it is more than 8 points
Click to expand...
8lade said:
Yep you can't just give those at the bottom maximum points and those already safe none, it doesn't work like that.
Click to expand...

Fair point, but how does it work?

If I were looking to have a bet on Norwich I would do this;

Points per game 18/26 = 0.692

Extrapolate 0.692*38 = 26.307

Looking at their remaining fixtures I would say that’s a reasonable number so my tenner would go on Norwich to get less than 30 points this season.

But that’s probability, to work out points for safety we have to look at mathematical possibility

Points remaining 3*12 = 36

Possible maximum points 36+18 = 54, so it is mathematically possible for Norwich to end the season on 54 points.

Looking at the minimum number of points needed to avoid relegation we have to set up the worst possible scenario for us (not winning again this season) and the best possible scenario for the teams below us (winning everything).

When all the remaining fixtures are played out you end up with improbable results like Norwich beating Man City, but that’s possible and it would be a bad result for us, so it goes into the results set to work out the final table.

I’m looking to work out an algorithm that can be written into software so you choose a team, click a button and see the number of points that team requires too be mathematically safe.

Last time I came up with 8 points for us, but Silent disputes that, so I’m going to run through the remaining fixtures following his suggestion and see what comes out.

It won’t be today because I’ve got more important things to do, like getting to BDTBL and watching the Blades give Brighton a pasting.

But any suggestions on how this works would be useful.

That’s fucking bollocks, isn’t helpful.

That’s fucking bollocks because ……… is

Thanks all UTB.
 
BladeByChoice said:
Fair point, but how does it work?

If I were looking to have a bet on Norwich I would do this;

Points per game 18/26 = 0.692

Extrapolate 0.692*38 = 26.307

Looking at their remaining fixtures I would say that’s a reasonable number so my tenner would go on Norwich to get less than 30 points this season.

But that’s probability, to work out points for safety we have to look at mathematical possibility

Points remaining 3*12 = 36

Possible maximum points 36+18 = 54, so it is mathematically possible for Norwich to end the season on 54 points.

Looking at the minimum number of points needed to avoid relegation we have to set up the worst possible scenario for us (not winning again this season) and the best possible scenario for the teams below us (winning everything).

When all the remaining fixtures are played out you end up with improbable results like Norwich beating Man City, but that’s possible and it would be a bad result for us, so it goes into the results set to work out the final table.

I’m looking to work out an algorithm that can be written into software so you choose a team, click a button and see the number of points that team requires too be mathematically safe.

Last time I came up with 8 points for us, but Silent disputes that, so I’m going to run through the remaining fixtures following his suggestion and see what comes out.

It won’t be today because I’ve got more important things to do, like getting to BDTBL and watching the Blades give Brighton a pasting.

But any suggestions on how this works would be useful.

That’s fucking bollocks, isn’t helpful.

That’s fucking bollocks because ……… is

Thanks all UTB.
Click to expand...
Am still sure that at present it is still mathematically possible to get relegated with more than 50 points especially if Norwich and Watford (or West Ham) also Liverpool and Man City lose their remaining games. I think the target of "more than 50 points" will remain the same even after all this weekend fixtures are completed.
 
8lade said:
Yep you can't just give those at the bottom maximum points and those already safe none, it doesn't work like that.
Click to expand...
Also the teams at the bottom all play each other at some point, it’s not possible for both teams to get maximum points.
 
BladeByChoice said:
Fair point, but how does it work?
Click to expand...
the problem is (unless I've missed something), the algorithm if you can even call it that will be enormously clumsy however software being what it is once it's written it'll be done for future reference. I'll just pause for a moment here to inform the world that although I know how to write software, in practice I was absolutely terrible at it so it may be worth finding somebody who has an aptitude for it. Mind you now I'm thinking about it there may be a way...I'll come back to this.
 
BladeByChoice said:
Fair point, but how does it work?

If I were looking to have a bet on Norwich I would do this;

Points per game 18/26 = 0.692

Extrapolate 0.692*38 = 26.307

Looking at their remaining fixtures I would say that’s a reasonable number so my tenner would go on Norwich to get less than 30 points this season.

But that’s probability, to work out points for safety we have to look at mathematical possibility

Points remaining 3*12 = 36

Possible maximum points 36+18 = 54, so it is mathematically possible for Norwich to end the season on 54 points.

Looking at the minimum number of points needed to avoid relegation we have to set up the worst possible scenario for us (not winning again this season) and the best possible scenario for the teams below us (winning everything).

When all the remaining fixtures are played out you end up with improbable results like Norwich beating Man City, but that’s possible and it would be a bad result for us, so it goes into the results set to work out the final table.

I’m looking to work out an algorithm that can be written into software so you choose a team, click a button and see the number of points that team requires too be mathematically safe.

Last time I came up with 8 points for us, but Silent disputes that, so I’m going to run through the remaining fixtures following his suggestion and see what comes out.

It won’t be today because I’ve got more important things to do, like getting to BDTBL and watching the Blades give Brighton a pasting.

But any suggestions on how this works would be useful.

That’s fucking bollocks, isn’t helpful.

That’s fucking bollocks because ……… is

Thanks all UTB.
Click to expand...
You seem to be working on avoiding finishing bottom, rather than guaranteeing at least 17th.

To find the highest possible points for 17th/18th you need to maximise the points for teams finishing 18th and above who aren’t already safe, so the model would require Norwich and either Watford or West Ham, plus Liverpool, to lose to all the other teams.

I explained my method in the opening post and shared an example possible league table where 17th and 18th would finish on 55, meaning 56 is currently needed to guarantee safety.

You wouldn’t need to calculate per team, as it’s the same number for everyone, except possibly the bottom two but I couldn’t be bothered to work that out as it doesn’t apply to us.
 
8lade said:
the problem is (unless I've missed something), the algorithm if you can even call it that will be enormously clumsy however software being what it is once it's written it'll be done for future reference. I'll just pause for a moment here to inform the world that although I know how to write software, in practice I was absolutely terrible at it so it may be worth finding somebody who has an aptitude for it. Mind you now I'm thinking about it there may be a way...I'll come back to this.
Click to expand...
Since Football Manager introduced the possible finish range for a team in their league I’ve wanted to know how complex that algorithm is.

For the first half of the season it would be 1st-20th, but after that it becomes dependent on what the remaining fixtures are as well as max available points. I guess there’s a formula to be calculated considering current points vs current points of the remaining fixtures which I may try to figure out some time.
 
8lade said:
the problem is (unless I've missed something), the algorithm if you can even call it that will be enormously clumsy however software being what it is once it's written it'll be done for future reference. I'll just pause for a moment here to inform the world that although I know how to write software, in practice I was absolutely terrible at it so it may be worth finding somebody who has an aptitude for it. Mind you now I'm thinking about it there may be a way...I'll come back to this.
Click to expand...
Punk Blade said:
You seem to be working on avoiding finishing bottom, rather than guaranteeing at least 17th.

To find the highest possible points for 17th/18th you need to maximise the points for teams finishing 18th and above who aren’t already safe, so the model would require Norwich and either Watford or West Ham, plus Liverpool, to lose to all the other teams.

I explained my method in the opening post and shared an example possible league table where 17th and 18th would finish on 55, meaning 56 is currently needed to guarantee safety.

You wouldn’t need to calculate per team, as it’s the same number for everyone, except possibly the bottom two but I couldn’t be bothered to work that out as it doesn’t apply to us.
Click to expand...

Software developer since 1997, started way back using COBOL and Fortran.

But we are not anywhere close to software development yet.

Basic software design. First get the algorithm down using pen and paper, napkin back of a fag packet, anything, before writing a line of code. Then you know the best language to translate the algorithm into a working application.

My current algorithm is:

Start with the club we are interested in (in this case Sheffield United).
Play out their remaining fixtures having them lose every game.
Resort the table.
Play out the remaining fixtures of the bottom club, having them win every game.
Resort the table.
Repeat this process until all the remaining fixtures for the season have been played out.

The resulting table for this process for match week 26 is in my post #89

To get the final result look at the team in 17th and calculate 46-39+1 = 8

I know this isn’t right yet because if I start with 18th place (the highest position to be relegated) I get different numbers. Sometimes higher and sometimes lower, so its not consistent.

If you can improve on this, please do, thanks.
 
Ok I've just wasted the best part of an hour to discover the safe line before a ball is kicked is 64 pts (or 63 and superior GD) and this slowly falls as games are played.
I've calculated the safe line is now 54 pts but there is still a margin of error which you can't remove without looking at each individual fixture.
 
I need to stop thinking about this now but I made an error in the calculation of 54 pts, it's 44 pts 45 if you want to remove the need to look at GD and there is still a margin of error.
 

BladeByChoice said:
Software developer since 1997, started way back using COBOL and Fortran.

But we are not anywhere close to software development yet.

Basic software design. First get the algorithm down using pen and paper, napkin back of a fag packet, anything, before writing a line of code. Then you know the best language to translate the algorithm into a working application.

My current algorithm is:

Start with the club we are interested in (in this case Sheffield United).
Play out their remaining fixtures having them lose every game.
Resort the table.
Play out the remaining fixtures of the bottom club, having them win every game.
Resort the table.
Repeat this process until all the remaining fixtures for the season have been played out.

The resulting table for this process for match week 26 is in my post #89

To get the final result look at the team in 17th and calculate 46-39+1 = 8

I know this isn’t right yet because if I start with 18th place (the highest position to be relegated) I get different numbers. Sometimes higher and sometimes lower, so its not consistent.

If you can improve on this, please do, thanks.
Click to expand...
Take the two teams with the lowest possible maximum total points and have them lose all their remaining games (except if they still have to play each other, then make it a draw).

Calculate the third lowest possible maximum points, and have any teams already above this total lose all their remaining games.

Then iteratively have third bottom beat the highest team not already safe and re-sort the table.

The challenge then comes when the fixture you need next isn’t left to play, and you may find that the highest total you can get for 18th isn’t as many as a team that you didn’t think was safe yet already has, based on the remaining fixtures, turns out they are. So you make them lose every game and start the iterations again.

It turned into a rather manual re-jig of remaining results, including a draw or two. It’s basically about redistribution of points among those from 18th up who aren’t yet safe, so the total of the points differences between 18th and each team above 18th but not safe is less than the number of teams on the same points as 18th, meaning there aren’t enough points left to redistribute to get 18th to a higher total.

Current max total for 18th = 55
Total for guaranteed survival =56
Total required for Blades right now = 56 - 39 = 17
 
According to my calculations

1. Liverpool 38 76
2. Man City 38 64
3. Leicester 38 56

Arsenal, Palace, Everton, Brighton, Burnley, Bournemouth, Villa, Southampton, Wolves 38 55


West Ham , Newcastle, Chelsea, Man U, Spurs 38 54

18. Blades 38 53
19. Watford 38 31
20. Norwich 38 20

So the magic number is 55 points, I think
 
The chances of 14 clubs finishing 1 point apart from this point are roughly 230007895 to one to get those remaining 221 results to fall that way
Plus 14 of our squad would need to have succumbed to the corona virus
 
When west ham lose at Liverpool tonight that means the blades have made another 1 point to the gap to possible relegation . Its now 16 points to the bottom 4 . Thats 6 Games .none of the bottom 4 will win more than half of their last 11 games. Will gladly take bets from anyone who think otherwise and will give 50 to one to anyone stupid enough that 2 of them could manage it
 
Isn't it time this tread was closed to further replies we were safe weeks ago when we reached 36pts ffs. It wasn't the case then but recent results now have the bottom five teams averaging less than a point per game. Carry this form on and best placed Bournemouth finish the season on 37pts, even 36pts will see one team safe as a lot were saying at the start of the season. It really is a pointless pass time trying to work out scenario's where we could still be relegated it could drive you to madness, it is not going to happen as it wasn't going to happen weeks ago.
 
metalblade said:
Isn't it time this tread was closed to further replies we were safe weeks ago when we reached 36pts ffs....
Click to expand...

My hope is that when all 38 games have been played, those posters who have been earnestly discussing this will re-visit this thread and realise how completely ridiculous it was.

Thereby saving the board from similar waste-of-time discussions next season.
 
cahill said:
This is an ironic thread ? Isn't it ? :confused:
Click to expand...

It's an English comprehension test. I'll let posters mark their own papers.

Punk Blade said:
There has been some debate recently around the difference between being mathematically safe (can no longer finish in the bottom three regardless of remaining results) and being effectively safe (we have more points now than the 18th placed team will have in the final table). This led to a few posters asking what are the actual required points for survival based on the current league table.
Click to expand...
Cassius Kray said:
I think it's an interesting thing to track, although it'd different from saying "how many points will we need to stay up".
Click to expand...
blade67 said:
I'm not sure anyone really thinks we aren't safe.

The OP is responding to the question posed by some on other threads about mathematical certainty. This must be really difficult to work out and a bit pointless, but it seems he's not the only one, as 2 or 3 have come to the same number of around 56 points - this week. It will have changed by next week.

Obviously, it won't need 56 points as the shit teams at the bottom will keep getting 1 point a game on average.

I'm interested to see when it actually becomes impossible
Click to expand...
Greasy Chap Butty said:
Btw I think we are coming from it from the same angle i.e we are safe, let's do some nerdy stuff....
Click to expand...
BushBlade said:
It’s pointless from a practical perspective but it seems we’ve got quite a few fans of maths on here. Applying theories to real life situations makes it more ‘interesting’. It’s a bit complicated for me but I don’t mind being a spectator.
Click to expand...
BladeByChoice said:
Its an exercise in futility by those of us who like doing sums in our spare time.
Click to expand...
Punk Blade said:
I’m going to keep updating this just to annoy the people who aren’t interested in it.
Click to expand...
Silent Blade said:
I agree, BladeByChoice and others were just discussing the mathematical possibilities
Click to expand...
 
Once again; if a thread offends you pluck out your eyes or just don't read it, it's all the same to those of us who like to speculate on things. I've quite enjoyed discussing this and shall continue to do so however nerd/ridiculous/whatever it makes me look.
 
8lade said:
Once again; if a thread offends you pluck out your eyes or just don't read it, it's all the same to those of us who like to speculate on things. I've quite enjoyed discussing this and shall continue to do so however nerd/ridiculous/whatever it makes me look.
Click to expand...
The thread doesn't offend anyone well I don't think it does but it is a pointless exercise we are not going down this season it does not matter how long anyone discusses the matter it will have the same outcome. This thread is the written equivalent of self harming.
 

8lade said:
Ok I've just wasted the best part of an hour to discover the safe line before a ball is kicked is 64 pts (or 63 and superior GD) and this slowly falls as games are played.
I've calculated the safe line is now 54 pts but there is still a margin of error which you can't remove without looking at each individual fixture.
Click to expand...
A number of fans here playing with numbers and trying to be alarmist or having fun ? having a laugh is my opinion!.
I estimate without doing maths 43/45 is possibly the biggest total a team this season could be relegated.For that to happen the top 10 teams would have to fall away dramatically ( collapse) good for a laugh.
 
You must log in or register to reply here.

All advertisments are hidden for logged in members, why not log in/register?

All advertisments are hidden for logged in members, why not log in/register?

Back
Top Bottom