Rolling Premier League table

How does a "rolling league table" work?

- Take the nine 2013/14 games played for Saints so far.

- Look at how they compare with the corresponding games in 2012/13.

- For example Saints lost at Liverpool last season 1-0, but this year Saints won at Anfield so Saints are given +3 points.

- Add all them up and Saints are +14 points better off this season than from the 9 corresponding games last season.

What about the promoted/relegated teams?

- You swap the results of the promoted sides with that of last seasons relegated sides.

- Swapping Wigan for Cardiff, Reading for Hull and QPR for Palace

How does the projection part work?

- Man Utd got 89 points in 2012/13, they are -8 points down on the corresponding fixtures in 2013/14. Which gives them a projection of 81 points.

- Saints had 41 points in 2012/13, they are +14 points up on the corresponding fixtures in 2013/14. Which gives them a projection of 55 points.

Although it has its flaws as a projection, I quite like it and as the season progresses the projection should become more accurate after each game.

Why are United projected 81 points? Or am I reading it wrong?

Why are United projected 81 points? Or am I reading it wrong?

They got 89 points in 2012/13, they are -8 points down on the corresponding fixtures in 2013/14. Which gives them a projection of 81 points.

Saints on the other hand had 41 points in 2012/13 and have 14 more points in 2013/14 from corresponding fixtures so their current projection is for 55 points.

They got 89 points in 2012/13, they are -8 points down on the corresponding fixtures in 2013/14.

Ah ok, so it's based on teams getting the same results in their remaining fixtures. That part is a bit meaningless then really.

Ah ok, so it's based on teams getting the same results in their remaining fixtures. That part is a bit meaningless then really.

Well its a projection based on similar data from last season. I think it is quite a good projection and becomes more accurate with each result. Hence the "rolling" part of the thread title.

Why are United projected 81 points? Or am I reading it wrong?

Who knows how to read that? I certainly can't see how it is concluded that those teams will finish in those positions based on the preceding stats.

They got 89 points in 2012/13, they are -8 points down on the corresponding fixtures in 2013/14. Which gives them a projection of 81 points.

 

Saints on the other hand had 41 points in 2012/13 and have 14 more points in 2013/14 from corresponding fixtures so their current projection is for 55 points.

so what is being projected is as if they would now pick up as many points as they did in the last 29 games last season...yeah right. Unless the green table is last seasons final 29 games and this season's first 9.

Assumes teams who have been very different to last year, up to now (eg Saints, Man U etc) will revert to performing identically to last season starting from today. Hard to forecast without a crystal ball but this is meaningless

Interesting that Arsenal got 27 points from their corresponding fixtures last season!

So it's if the league ran from exactly 12 months ago?

So it's if the league ran from exactly 12 months ago?

No.

We seem to have done really well compared to the same fixtures last season, but it will be hard to replicate the home wins against Liverpool, Man C, Chelsea.

It's quite interesting viewing and I could see the final table looking similar, but it doesn't really extend the current level of performance over the season. Mlg, if you have time what would it look like if you did that, so to explain if manure are -8 over 9 games, they'd be -32 over 36 games (times each by 4). Completely pointless speculation but fun all the same.

We seem to have done really well compared to the same fixtures last season, but it will be hard to replicate the home wins against Liverpool, Man C, Chelsea.

Ah but we still have Cardiff (ie wigan) at home to catch up a few points should we not win those 3, you could say that WBA won't tonk us at home either on this year's form so that's more points we can make up fairly easily,idem norwich at home unless we play Gazza to gift them a goal or two that is. Just last year's form plus our good start plus those 3 games would see us on somewhere around 60+ points wthout improving on results like Sunderland and Fulham away.

No.

I believe i've got it now, thanks to MLG for spelling things out clearer in the key beneath.

A 'rolling' 12 months from point to point would be interesting for me. Can understand the logic of doing it fixture to fixture mind.

A great analysis. I can see the logic but we all know that football is never that easy to predict, which is what makes it such an enthralling sport to follow.

If it shows nothing else then it shows how much the rest of the league has reeled in ManU and how we have surged out the pack of the lower teams and are headed up the table to challenge for that top 7, 6, etc.

Be interesting to keep this up to date so we can see ManU be "knocked off their perch" so to speak

Although it has its flaws as a projection I quite like it and as the season progresses the projection should become more accurate after each game.

a bit like the actual league table.

pointless statto porn.

a bit like the actual league table.

 

pointless statto porn.

But unlike the actual league table it projects what things might be like at the finish, using previous data of similar events last year to fill in the blanks of the remaining 29 games in combination with what we know from the opening 9 games.

All harmless fun. Cheer up.

I think it's a pretty good measure of current form rather than a predictor of the final table.

But unlike the actual league table it projects what things might be like at the finish, using previous data of similar events last year to fill in the blanks of the remaining 29 games in combination with what we know from the opening 9 games.

 

All harmless fun. Cheer up.

I think I'd rather base them on similar events from this year, to be honest. Smaller sample but probably going to be more accurate with nearly a quarter of the season gone.

a bit like the actual league table.

 

pointless statto porn.

Indeed, I think if someone shows the points total for each team across many seasons that will indicate the variability and see if this type of projection is in any way indicative or jerk off material.

It's nice to see just how much worse than the TERRIBLE QPR that Palace are though.

How would the league table look if it only included goals scored between the 23rd and 69th minute by players born in any contient other than Europe that dont have the letters T, Q, R, L, I, P or N in their first or surnames?

It's nice to see just how much worse than the TERRIBLE QPR that Palace are though.

They might take Derby County's record of 11 points in 2007/08 at this rate.

It's nice to see just how much worse than the TERRIBLE QPR that Palace are though.

but after the 1st 9 games last season QPR actually had less points than Palace do today. 3 for QPR and 4 for Palace, mind you QPR had had some pretty rough fixtures by then last season, Citeh,Chelsea etc; They'd not won a single game in 9 which is why Hughes got sacked I guess. So there are 2 ways of looking at it from Palace's point of view, either they're roughly the same as QPR and are bound to go back down or that they have the same number of points as we did from 10 games last season and we stayed up.

I think I'd rather base them on similar events from this year, to be honest. Smaller sample but probably going to be more accurate with nearly a quarter of the season gone.

who's going to volunteer to apply this system to each of the last ten premier league seasons and compare the projection after 9 games with the final tables?

who's going to volunteer to apply this system to each of the last ten premier league seasons and compare the projection after 9 games with the final tables?

And update it every week.

also i've no idea why I quoted the9

Im confused

Hard to believe that from the corresponding fixtures last year we didn't get a single win - just 4 draws and 5 defeats!

Im confused

About anything in particular?

Good to see. I like that Arsenal are actually doing worse than they did last year.

Good to see. I like that Arsenal are actually doing worse than they did last year.

good point, that would in fact mean that against the 9 opponents they've had so far they won them all those games last season. Whereas this season they got beaten at home by Villa and only drew with West Brom away. The test for the Arse will be when they get their usual mid-season injury crisis as they would seem to have tougher opponents mid term this season.

I like the like-for-like result comparison, but the "projection" is meaningless.

We seem to have done really well compared to the same fixtures last season, but it will be hard to replicate the home wins against Liverpool, Man C, Chelsea.

Yeh but we also lost at home to Wigan and QPR for example to catch up points.

Who knows how to read that? I certainly can't see how it is concluded that those teams will finish in those positions based on the preceding stats.

The most it can ever be a realistic stat will be when we have played all teams once at the half way stage (even though we are unlikely to have played everyone home or away exactly as last season.) And we have the toughest part of the half yearly programme of matches to come, including against some of the top clubs. Also we have been lucky with injuries to date and it would not be unrealistic to expect our fast tempo style to eventually take its toll (but I hope not). So probably (imho) an eigth end of season position is not an unrealistic proposition. We may well climb the table to even higher than 3rd for short spells but I still consider eighth the benchmark to beat.

I'd like to think that we will end up on 76 points this season.....based on 2 points per game season to date

Sent from my GT-N5110 using Tapatalk

Of the 9 fixtures, five -liverpool (a), fulham (h), wba (a), swansea (h) and sunderland (h)- were under NA. This season, we've bettered all those results with a +11 point differential.

About anything in particular?

I found that interesting. Thanks for posting.

I like the like-for-like result comparison, but the "projection" is meaningless.

It's not really a projection though is it? Using that word is a bit of a red herring. It's the actual points earned in the most recent 38 games played against our 19 opponents once each at home and away. So every week that figure is subject to change as the newest game gets added but the previous matching fixture gets removed.

Taken as a snapshot it's not particularly relevant but followed weekly it'll give a good insight to who is improving (that'll be us then) and who's going the other way.

It's not really a projection though is it? Using that word is a bit of a red herring. It's the actual points earned in the most recent 38 games played against our 19 opponents once each at home and away. So every week that figure is subject to change as the newest game gets added but the previous matching fixture gets removed.

Taken as a snapshot it's not particularly relevant but followed weekly it'll give a good insight to who is improving (that'll be us then) and who's going the other way.

No that's wrong. The next fixture is vs stoke and we drew 3-3 last year, so if we win our "projection" goes up to 57, if we draw it stays at 55 and if we lose it drops to 24. As the season goes on, it gets more and more accurate.

Damn, can't edit, 24 should read 54...sorry.

So despite their stuttering start to the season, Man United are going to win the league anyway, because of all those games they won last year under their previous manager. Well done them. What a fantastic projection.

Statto nonsense almost biblical in its pointlessness.

Lol and now I have re-read your post wurzel and you say the same as me, sorry, just turned 40, brain misfiring...carry on...

Interesting to see comparisons with equivalent fixtures from last year.

The projection for end-of-season points tally is absurd though.

I think even if we won 15 games in a row, we'd still be projected to finish only 6th or 7th.

If Man Utd lost their first 23 games, they'd still be projected to stay up comfortably.

Which website is this from or is it generated by the OP?

Like for like comparisons is interesting but as others say projection is not.

As we are averaging 4.5 times as many points as last season the projection shpuld show we get 184 pts this season which I think might beat the record.

I think would be better to just have points after same number of games as well regardless of opposition although meaningless for promoted sides.

Same bores ripping into what is a fair discussion topic. Fck knows this place is repetitive enough (as are some posters) so anything like this is welcome.

Thanks MLG.

It is standard business practice to measure sales or margins on a rolling 12 months or moving average total. It has the advantage of making trends far easier to track especially in sales where particular months have traditionally higher or lower sales.

I had been keeping the same record for Saints this year to see how we were doing against my prediction of 55 points this season. At this rate I will be well out!

I think this chart has far more merit than people are giving it. The 'projection' isn't accurate (no projection ever is), but it is better than many considering it is based upon your last 1-2 results against each team. As such this is a bit like saying this is what will happen if every team reverts to the form of last season.

From a saints perspective that's pretty impressive, as half of our last season we were in the relegation zone. So essentially this chart tells us where we might end up if our recent form flags, and we spend 10 games playing like relegation candidates. In other words, because we've exceeded last seasons form, to finish in the 'projected' final position would actually need quite a collapse in our current form. To me that's both interesting and reassuring.

I think this chart has far more merit than people are giving it. The 'projection' isn't accurate (no projection ever is), but it is better than many considering it is based upon your last 1-2 results against each team. As such this is a bit like saying this is what will happen if every team reverts to the form of last season.

 

From a saints perspective that's pretty impressive, as half of our last season we were in the relegation zone. So essentially this chart tells us where we might end up if our recent form flags, and we spend 10 games playing like relegation candidates. In other words, because we've exceeded last seasons form, to finish in the 'projected' final position would actually need quite a collapse in our current form. To me that's both interesting and reassuring.

I come on this forum to read repetitive over-reactionary personal jibes, not this rational and pragmatic nonsense thank you very much! FFS.

Thanks MLG.

 

It is standard business practice to measure sales or margins on a rolling 12 months or moving average total. It has the advantage of making trends far easier to track especially in sales where particular months have traditionally higher or lower sales.

Except this is not a Moving/Rolling Average. That would be simply the last 38 games.