The Alternative Premier League Table 2014-2015

[ERO]

I disagree. After 25 matches, the table doesn't lie (beyond mere margins).

Small contradiction there. Those margins are also what this alternative table tries to reduce. If the table doesn't lie, then why don't we just wrap it up now rather than play 13 more matches?

It doesn't really matter if not all teams have met the exact same opponent the same amount of times.

Why not? Do you not think that Arsenal's preceived easier run in gives them an advantge? Would you not swap remaining matches left if you had the chance?

There are so many other variables that influence the standings beyond the name of the opponents you've met. The present table is the best at showing how good a club is performing relative to all other opponents.

Correct. And that is by definition what the league table does. What this alternative table tries to do is predict the outcome after 38 matches.


Say we are 1 point ahead of Arsenal going into the last round. We're playing Chelsea away, Arsenal is playing Leicester at home. The league table that according to your argument definitely doesn't lie after 37 matches has us finishing above Arsenal. This alternative table has us finishing below. I know which one I would back.

The alternative table merely puts probability weights on different outcomes rather than assuming everyone will do exactly the same for the remainder of the season - opposition irrelevant. It is harder to amass points away to the top 13 sides in the league than other matches - I don't understand who will argue that.

 

[Everlasting Seconds]

Correct. And that is by definition what the league table does. What this alternative table tries to do is predict the outcome after 38 matches.

But the question is, does it do any prediction better than the real thing, the actual one-and-only table? I think not. The present alternative table will most likely differ from the final table as much, not significantly less, than the present actual table will differ from the final table. Why? Scientifically speaking, the population making up the foundation for the alternative table is probably at the end of the day the same population as the one being the foundation for the actual table. Now, if the two populations are too similar or too alike (as I deem they are), it makes sense that the two tables will vary so little that one of them (the alternative one) becomes redundant, rather than adding any new layer of valuable information. You can predict the final table now based on the present actual table, and you will be as close to the final truth as if predicting using the alternative table.

 

[225]

Say we are 1 point ahead of Arsenal going into the last round. We're playing Chelsea away, Arsenal is playing Leicester at home. The league table that according to your argument definitely doesn't lie after 37 matches has us finishing above Arsenal. This alternative table has us finishing below. I know which one I would back

Arsenal have struggled against Leicester this season. As previously mentioned, the beauty of the PL is such that you can't ever be certain that 'big team beats little team' or 'top team beats relegation zone team'.

The 'alternative' table is not any more accurate, it just has a bias towards teams higher in the table and doesn't take into account the beautiful variance that actually exists.

 

[beats1]

I don't really use it, its good to see where we are but if you are in the top 4 race, its pointless as the objective for each match would different than that of the objective for a top 4 race.
Here is the updated graph for those who want it

Currently Arsenal top then Southampton followed by united, us then Liverpool.

 

[Geyzer Soze]

This is the most confusing thread on SC... Except for the Dibby one for anyone who joined after he left.

At least the Dibby one had interesting 'research' material

 

[npearl4spurs]

I just updated it for whoever was interested. If anyone wanted to take the info and use it to project the top 4 race or something...

It is obvious we needed to perform better at home this year as we have been one of the best away teams in the league, but I'm sure none of our fans needed an alternative table to realize that.

 

[BK007]

I just updated it for whoever was interested. If anyone wanted to take the info and use it to project the top 4 race or something...

It is obvious we needed to perform better at home this year as we have been one of the best away teams in the league, but I'm sure none of our fans needed an alternative table to realize that.

I pointed it out because you are clearly doing something wrong if it matches up exactly with the PL table.

The chart beats has posted with Man United slumming it with us and Liverpool is what the table was "supposed" to look like if you follow the link to the Liverpool forum's table.

 

[npearl4spurs]

I pointed it out because you are clearly doing something wrong if it matches up exactly with the PL table.

The chart beats has posted with Man United slumming it with us and Liverpool is what the table was "supposed" to look like if you follow the link to the Liverpool forum's table.

Okay thanks for pointing that out.

I'm at work now and can't take the time to try to be helpful so maybe you could help show me where I went wrong and what it is "supposed" to be?

 

[ERO]

To revisit this thread it's interesting to note that bar overestimating Southampton - as one could expect - the alternative table got it spot on position wise, and gave a much better prediction than the league table that never lies after 25 rounds.