Topic: The Alternative Premier League Table 2016-17

[Fluke]

It's actually fairly unusual for a lot of teams to swap from par one to par three over the course of a full season. The trad table at the end of the previous season is probably still more of a realistic guide than the table six weeks into the new campaign.
It's a lot of extra work just to offset two or at most four shifts.

Fair enough. It was a thought I had.

[Frizzo]

Wait, I think I might get what you are saying.  Are you saying DURING a round, there will be differences because team A might be 7th but after another team wins they become 8th in the par rankings and change from a par 1 away to a par 3, before another team plays them that round?

I guess the way to account for that is to lock the table until the game week matches are all complete?

No, I mean a team could theoretically play 4 or 5 different teams at a time when they're a par 1, and due to the rolling form another team could play all of those same teams at a time when they're a par 3.

[CraigDS]

An idea I had a few years ago was having the APLT pars be rolling live. By that I mean, rather than final position from the previous season, taking the 38 previous rounds (e.g. Round 6 last year until Round 5 this year).  Calculating the league position based on those results, and applying pars on a rolling basis, week to week.

But you'd have teams who have played a different mix of 38 teams during those rounds compared to anyone else.

Using a whole season every side has played the same games as the others.

[Nessy76]

Fair enough. It was a thought I had.

Yes, I can see the logic behind it, but I checked last year how often teams really do shift and it's not that common. Par 3 sides are more likely to be relegated than to become par one sides, and newly promoted teams hardly ever make the top thirteen spots.

I also like the fact that you can track the graphs as the season progresses, which you can't if they change every week.

[Nessy76]

No, I mean a team could theoretically play 4 or 5 different teams at a time when they're a par 1, and due to the rolling form another team could play all of those same teams at a time when they're a par 3.

I think Fluke is proposing to redraw the whole graph every week to take an up to date view based on current league position.

[Frizzo]

I think Fluke is proposing to redraw the whole graph every week to take an up to date view based on current league position.

Then couldn't some teams have played Everton 3 times for example?

[markthescouser]

No mate, at the minute, everyone has 12 par ones and 26 par threes.

A rolling table where everyone changed position each week would mean some teams end the season having played say, 15 par 1 games whilst another team could have only played 9.

To get around this issue you would have to make the changes in position each week retrospectively to each previous week too.

I believe it would work and probably give a slightly more accurate model, but would be a lot of work for someone, for what I imagine is only a couple of points here and there for some teams.

[Nessy76]

Then couldn't some teams have played Everton 3 times for example?

While that would obviously be a huge advantage to anyone, it's only the pars that would be revised, I think the table itself would still be based on the current season alone, unless I'm reading it wrong.

[Frizzo]

While that would obviously be a huge advantage to anyone...

Bravo, sir.

[kcbworth]

I think Fluke is proposing to redraw the whole graph every week to take an up to date view based on current league position.

Might make sense once a clear pecking order has been established. In the first ~10 rnds, it's likely too variable?

[SP]

If you are automating the generation of the graphs, making assignment of which teams are par 1 the first step of that process is not too onerous. In terms of accuracy, you would probably have to use a ppg table rather than the actual table to reduce the effects of the variable number of games played.

Of course if you were being grandiose, you could use an updating full table - using the matches from the current season, filling in the gaps with the results from the previous season - using relegation/promotion team substitution where appropriate. That would factor in the current form, and still use a full season worth of results.

Mind you, I could not be arsed to set that up, and I my "being arsed" threshold is quite awesome. The current approach is simple, yet good enough.

[Fluke]

SP, that's exactly what I had in mind.  Nessy and Mark described it well.

I actually think it wouldn't be too hard to set up, if you had a feed providing the results. The problem is I have been especially lazy... I might try it during this season though, just to see how it looks.

[Banquo's Ghost]

My own view is that in these discussions which happen every year, people are taking this thread way too seriously.

The only table that matters is the real one. The APLT provides a perspective on the positions in the real table that can help understand the challenges ahead. Trying to come up with alternative and complex 'weighting' misses the point and plays into the ridicule that we somehow think the APLT is more meaningful than the actual standings.

The APLT's strength is its simplicity. It provides a useful discussion point. That's all.

[riismeister]

Hey, some of us like math and spreadsheets and stuff.

[Deo]

The APLT is about judging a team's fixture difficulty relative to their position on the real table. The APLT doesn't make any sense without the real table, so the comparisons really are pointless.

We're rightly where we need to be because we lost to god damn Burnley who scored two of their grant total of three goals this season against us.

City on the other hand, despite their lofty +6, haven't really been tested so far. Their only difficult fixture has been vs United, while the +2's they gained vs Stoke and Swansea are questionable because those two teams are awful this season. Their true test will come this weekend vs Spurs, while our true test is consistency, especially vs the smaller teams.

[scalatore]

My own view is that in these discussions which happen every year, people are taking this thread way too seriously.

The only table that matters is the real one. The APLT provides a perspective on the positions in the real table that can help understand the challenges ahead. Trying to come up with alternative and complex 'weighting' misses the point and plays into the ridicule that we somehow think the APLT is more meaningful than the actual standings.

The APLT's strength is its simplicity. It provides a useful discussion point. That's all.

Right. I've been following the APLT threads for years because it's a very useful way to consider fixtures and results at certain stages of the season. Every year people come in wanting to make it more complicated. Just let it be simple!

[Prof]

The only way a rolling par would work without teams having different par totals would be if the pars are adjusted for matches already played if the rankings change.  Therefore anything you see happen in one week might mean something different the following week.  There's also the issue that league table rankings are influenced by the difficulty of fixtures to date.  There would also need to be an aplt to decide the rankings of teams to identify the pars which makes things even more complicated and unnecessarily so.

Therefore, I think it's a non-starter.

[drmick]

I'm not sure a rolling par offers much more insight than just a plain rolling points total.  Sky often do a "if the league began 1st January......" to highlight teams that have recently come into form, or have suffered massive drops in form.

If we were to do a "if the league began 38 games ago" wouldn't that likely show a similar ranking of teams to an APLT based on those 38 games?

[Hij]

My head hurts


[You may have an iPhone but you are still using the free version of Tapatalk and spamming us with that knowledge]

[RooiBefok]

APLT Full Table : Round 6

[Zimagic]

Great work on the APLT. Did someone reboot the results comparison thread too?

Loved that one for watching the progression from year to year.

[KurtVerbose]

If Nessy76 is doing a weekend preview I will very much look forward to reading - Nessy, please don't feel obligated!

We have a par 1 away to Swansea, Man City have a par 1 away to Spurs. The bad news is Spurs will be on a CL hangover and are without Kane.

[Nessy76]

If Nessy76 is doing a weekend preview I will very much look forward to reading - Nessy, please don't feel obligated!

We have a par 1 away to Swansea, Man City have a par 1 away to Spurs. The bad news is Spurs will be on a CL hangover and are without Kane.

Cheers, all depends how busy I get on Friday afternoon!

[rickardinho1]

Gameweek 7:

Friday
Everton vs Palace       (Par 3)

Saturday
Liverpool @ Swansea (Par 1)
Chelsea @ Hull           (Par 3)

Sunday
United vs Stoke          (Par 3)
Spurs vs City              (Par 3)
City @ Spurs               (Par 1)
Arsenal @ Burnley      (Par 3)

Great opportunity to gain ground on everyone as rivals will inevitably drop points.

[Hij]

We have a par 1 away to Swansea, Man City have a par 1 away to Spurs. The bad news is Spurs will be on a CL hangover and are without Kane.

City have got a tough away tie in Europe a day after Spurs

[Frizzo]

Cheers, all depends how busy I get on Friday afternoon!

Oh now the excuses start...

[Brian Blessed]

You've let the entire site down, Nessy. Shame.

[Fluke]

I think you guys are right as far as the value of the rolling par.  I was just curious to see if it painted a different picture, as later in the season it would be more reflective of each team's current form, rather than just their previous season's form.

Though as mentioned, it would definitely just be a snapshot, as the values of past matches would change as people changed par values. The entire graph would change.

(Though Nessy pointed out this doesn't happen very often, so the graph may not change that much).

By the way, I never suggested this as a replacement to the APLT, I love the APLT. I have just been curious to see what that might look like in conjunction with it.

[Redman0151]

What are Everton now, -7 already? Or is Bournemouth away par 1?

[ryatnalkar]

Fwiw I think the rolling Par is a really good idea. Someone mentioned that some team might have to face 15 PAR 1 vs some other team only 9 PAR 1... But isn't that true over the season? Some teams play orher teams  the right times? Or for another example, if Leicester really keep playing the way have so far and Chelsea play themselves towards the top 2, is it totally pointless to have Chelsea be par 1 and Leicester a PAR 3? Rolling PAR seems obvious improvement over current scheme.

[zabadoh]

A rolling par would be a lot of work.  Every team would have to have its own par ranking table because each team doesn't play the same opponent every week.  You can't roll the results on a temporal basis either because teams would be affected arbitrarily by the schedules

Sent from my SM-G386T1 using Tapatalk

[TepidT2O]

What are Everton now, -7 already? Or is Bournemouth away par 1?

they were added to the table...


Maybe time to remove them again

[Frizzo]

they were added to the table...


Maybe time to remove them again

It was far too early to add them.

[McrRed]

A rolling par would be a lot of work.  Every team would have to have its own par ranking table because each team doesn't play the same opponent every week.  You can't roll the results on a temporal basis either because teams would be affected arbitrarily by the schedules

Sent from my SM-G386T1 using Tapatalk

I was thinking about this last year and came up with a coefficient based on form from the last 6 games. It tied in well as a predictor, but, like most of the other alternative alternatives just isn't as elegant as the original.

[IanZG]

It was far too early to add them.

To be fair, they did look better than Leicester, not sure who I'd choose at this point to be the 7th team...

[Frizzo]

To be fair, they did look better than Leicester, not sure who I'd choose at this point to be the 7th team...

To be fair, it's Everton.

[Prof]

they were added to the table...


Maybe time to remove them again

Not in mine.  Only in JCB's which is dynamic and can select any teams wanted.

[Iska]

It was far too early to add them.

But if they hadn't been added, we wouldn't get to remove them.

[Frizzo]

But if they hadn't been added, we wouldn't get to remove them.

Fair play sir.

[Prof]

Fwiw I think the rolling Par is a really good idea. Someone mentioned that some team might have to face 15 PAR 1 vs some other team only 9 PAR 1... But isn't that true over the season? Some teams play orher teams  the right times? Or for another example, if Leicester really keep playing the way have so far and Chelsea play themselves towards the top 2, is it totally pointless to have Chelsea be par 1 and Leicester a PAR 3? Rolling PAR seems obvious improvement over current scheme.

If every team has the same par score (90), after 38 games, the aplt matches the real table.  If the pars are different, there might be different teams at the top of each table.