DanTheMan Posted 25 November 2011 Share Posted 25 November 2011 In previous threads I've said I don't like to get too bogged down in the stats - for example, win percentages don't mean a lot in the real world, etc etc. HOWEVER, whilst procrastinating at work, I have compiled a look at our form this season, and compared it to the historical promotion form for the Championship in years gone by. I'd argue these stats DO MATTER, as it compares our current form with long-term statistics, and it gives us an idea, even at a pre-christmas stage in the season, of how we are doing. A few notes on the above graph: Predicted Total: Based on our average points-per-game after a given result, then extrapolated over the 46 league games of the season. This gives us an idea of our current form/pace after any given match that we've played so far. 10yr Average/Best/Worst Case: These lines represent the average points required to achieve at least the 2nd auto place (i.e. 1pt above the team that finished 2nd that season) as well as the lowest (best case) and the highest (worst case) points required in the last 10 seasons to finish 2nd. Max Points: This is our maximum possible points total after a given result. Obviously not very useful (since when does any team win all there remaining games?!) but it does illustrate the amount of breathing room we have in order to still finish at least 2nd, as we'd ideally like to keep our max points above the 10year average line to ensure promotion! Anyway, I welcome a healthy discussion on these figures - hopefully people will agree that stats like this are much more useful than things like win percentages (I know I sparked a healthy debate in my last thread!) EDIT: It was pointed out that I'd been an idiot and worked out the Average/best/worst case lines wrong - I'd worked them out using the assumption that 1pt above the team that finished 3rd in a given season would mean we finished 2nd, but obviously then WE would finish 3rd! Graph now fixed - original is here: http://i.imgur.com/Bxxff.png for comparision! Link to comment Share on other sites More sharing options...
shade Posted 25 November 2011 Share Posted 25 November 2011 isn't the graph wrong? shouldn't the average/worst/best case lines be 1 point above the second place team that year, not third? by putting it 1 point above the third placed team that season, we would finish 3rd? Link to comment Share on other sites More sharing options...
Charl91 Posted 25 November 2011 Share Posted 25 November 2011 isn't the graph wrong? shouldn't the average/worst/best case lines be 1 point above the second place team that year, not third? by putting it 1 point above the third placed team that season, we would finish 3rd? You're right, but by 1point ahead of the 3rd placed team, I think he means that they're third after he puts us in front of them (i.e they were 2nd before, just seems worded badly). That was my interpretation anyway, could be wrong though. Edit: I guess I was wrong Either way though, a very good graph. Worrying that we've already used up about half of our "breathing room". Just out of interest, have you checked how the current top 2 are doing in comparison to the previous ones? Link to comment Share on other sites More sharing options...
DanTheMan Posted 25 November 2011 Author Share Posted 25 November 2011 isn't the graph wrong? shouldn't the average/worst/best case lines be 1 point above the second place team that year, not third? by putting it 1 point above the third placed team that season, we would finish 3rd? WHOOPS! Quite right - I am a grade A toolbox! Graph now fixed - the upshot is we now have even less room for manoeuvre! Link to comment Share on other sites More sharing options...
Hitesh Posted 25 November 2011 Share Posted 25 November 2011 Have you sent this to NP? Link to comment Share on other sites More sharing options...
burni Posted 25 November 2011 Share Posted 25 November 2011 nice graph took me a few looks to work out what exactly was going on but i get it now! i think this shows more than anything we need to be on a consistent run for most the remaining season to be in an automatic promtion. Be interesting to see this graph but for play offs rather than automatic promotion? just think this is more likely what were aiming at. Link to comment Share on other sites More sharing options...
DanTheMan Posted 25 November 2011 Author Share Posted 25 November 2011 As requested by a couple of people... Auto spot comparison with Southampton and West Ham: LCFC form for finishing in play-offs: Interesting stuff - looks like with their recent run, West Ham has pulled away a bit, I reckon we'd need a similar run in order to stand a chance of hitting the 2nd auto spot. Playoffs on the other hand look very possible if we can just get a bit more from the team and maybe have Pearson work some of his magic! Link to comment Share on other sites More sharing options...
woddyuk Posted 25 November 2011 Share Posted 25 November 2011 Can you help me budget my mortgage and utlilties too ??? ;-) Link to comment Share on other sites More sharing options...
Iwan is a Welshman Posted 25 November 2011 Share Posted 25 November 2011 Mate that's brilliant. Link to comment Share on other sites More sharing options...
John Matrix Posted 25 November 2011 Share Posted 25 November 2011 quality. love stuff like this. Link to comment Share on other sites More sharing options...
Les-TA-Jon Posted 25 November 2011 Share Posted 25 November 2011 Stats based opinion making is the way forward!!!!! Forgive if I'm being an idiot - but why is it that all the graphs have our worst case line above our best case line? Link to comment Share on other sites More sharing options...
Salieri Posted 25 November 2011 Share Posted 25 November 2011 Interesting analysis, appreciate your efforts. Link to comment Share on other sites More sharing options...
millzi Posted 26 November 2011 Share Posted 26 November 2011 Stats based opinion making is the way forward!!!!! Forgive if I'm being an idiot - but why is it that all the graphs have our worst case line above our best case line? It took me a while to figure it out too, but I eventually got it. In the best case scenario we would only need to get minimal amount of points to be promoted, whereas in the worst case scenario we would need to get more points, thus winning more games and less room to slip up. Hope that cleared things up for you P.S great graph Dan, looks very clean and professional, top work. Link to comment Share on other sites More sharing options...
Finnegan Posted 26 November 2011 Share Posted 26 November 2011 Kokopops just got erection. Link to comment Share on other sites More sharing options...
Alexikokopops Posted 26 November 2011 Share Posted 26 November 2011 Kokopops just got erection. Actually, I sat there and gave some thought about the forecasting method he used to predict our final points I like it very much, but I wonder if the method use means that the "predicted points" value won't be as reactive to the kind of changes that can take place in football (like the sacking of a manager etc.). We all know the way you can go on a run in football (especially in the Championship) and storm up the table. I'd be interested to see what the "predicted points" line looked like if you worked out the average points from the last 6 games and used that to extrapolate over the rest of the season. That way, if you had a bad start but suddenly started going on an awesome run (like Palace did that time under Dowie, for instance) your final points prediction wouldn't be biased by the poor start (which it would if you worked out the average points based on the whole season's games). It would naturally be more volatile than Dan's method. That said, my method above is still very simplistic. However, I'll be damned if I'm going to sit there and try and do a proper forecast on something as volatile as football. Now I've got an erection. Link to comment Share on other sites More sharing options...
DanTheMan Posted 26 November 2011 Author Share Posted 26 November 2011 There are definitely some better ways (including Alexikokopops' suggestion) for predicting our actual end of season tally, or showing form trends - but I think my graphs are more centred on showing our "pace" after a given game, with a simplistic extrapolation of our points at that point, relative to other teams whilst comparing these "pace" values with the long-term points tallys required for the playoffs or promotion, etc. That said, it seems like there is a fair bit of interest in these graphs - I am happy to update them throughout the rest of the season if people find it useful/interesting to keep up-to-date! What would be the preferred option? I could post to this thread after every 3-4 matches? Does this sound like a good idea? Link to comment Share on other sites More sharing options...
shade Posted 26 November 2011 Share Posted 26 November 2011 no, posting to it after every match seems like an even better idea! Link to comment Share on other sites More sharing options...
DanTheMan Posted 26 November 2011 Author Share Posted 26 November 2011 no, posting to it after every match seems like an even better idea! I can see this thread getting lengthy very quickly! Still don't think it'll give the Maynard thread a run for it's money though! Link to comment Share on other sites More sharing options...
Alexikokopops Posted 26 November 2011 Share Posted 26 November 2011 There are definitely some better ways (including Alexikokopops' suggestion) for predicting our actual end of season tally, or showing form trends - but I think my graphs are more centred on showing our "pace" after a given game, with a simplistic extrapolation of our points at that point, relative to other teams whilst comparing these "pace" values with the long-term points tallys required for the playoffs or promotion, etc. Definitely. The problem with my suggestion is that it can flit about quite a bit and lead to increased optimism/pessimism. The problem with yours is that as we get later on in the season, each result is going to do less and less to change the average (because of a larger sample size), and so could underestimate what we can do. That said, I think the addition of the max points line helps with that - while at this stage it's not that important, towards the end of the season I imagine it will be a lot more useful. Also, knowing the speed with which opinion can swing on this board, I favour your method That said, maybe a larger rolling average than just the six games would work? It would reduce the volatility of a six game average, but increase the ability for the prediction to react to things like a good run of form. I'm just musing, mind. Link to comment Share on other sites More sharing options...
Bob Weasel Fox Posted 26 November 2011 Share Posted 26 November 2011 all a bit graphic for me Link to comment Share on other sites More sharing options...
davieG Posted 31 December 2011 Share Posted 31 December 2011 Any chance of an update? Link to comment Share on other sites More sharing options...
DanTheMan Posted 5 January 2012 Author Share Posted 5 January 2012 Hi again all, I thought I'd post an update now that we are just over halfway through the season and in the FA cup break... Firstly, lets look at our own auto-promotion performance again: As any realistic (pessimistic?) fan would have already figured out, short of a miracle, our hopes of reaching the 2nd spot by season's end are pretty much scuppered. Our performance since Blackpool at home has not only left us 12pts adrift, but we have used up virtually all our "spare-room" with regards to the maximum possible points line as shown above. Once that line drops into the range of points needed to guarantee 2nd, the planets would need to align before we'd finish there. Obviously if the max points drops below the best-case scenario line, it's not even mathematically possible anymore... Next lets take a look at our play-off prospects: Much better possibility of finishing here by the end of the season. If we can achieve consistency and use the Palace win as a platform, I see no reason why we can't finishing in the play-offs. Obviously the last few games haven't helped the cause, but we still much more "spare-room" to play with looking again at our "max points" line Finally, (although you could argue it's not that relevant anymore) an update of our performance vs Southampton/West Ham: This clearly illustrates what would have been if we'd have found our feet after winning at home to Blackpool... up to this point, both the Saints and West Ham were pulling away and looking unstoppable as promotion favourites. Since then, in the last 6 games, they have dropped points and been inconsistent: Southampton Form: L-D-D-W-L-L (0.83 pts/game) West Ham Form: L-L-W-D-L-W (1.17 pts/game) i.e FAR FROM PROMOTION FORM! It is said, and the stats prove, that an average of 2 pts/game over the season will almost always get you promoted (finish on 92pts). Leicester's form goes like this: Upto and including Blackpool: 1.52 pts/game last 6 games (inc Palace win): 1.00 pts/game If our form had stayed consistent after the Blackpool match we would be on 38pts. Had we kicked on from there and found the magical 2pts/game form, we'd now be on 41pts and in 5th place. Despite our less than amazing start to the season, and our general lack of consistency in results, this shows what finding our consistency and upping our form would have done in December. As it stands play-offs is a realistic goal to set ourselves - if we can get to the playoff final, I hope by then we will have shown that we have the players, and the TEAM to go on and win to spend the 2012/13 season in the Premier League. Fingers crossed folks! I welcome your thoughts! Link to comment Share on other sites More sharing options...
Trav Le Bleu Posted 5 January 2012 Share Posted 5 January 2012 Wow, even Southampton and West Ham won't get promoted! Seriously, nice analysis. I think our only hope is that nobody seems to want to get promoted this year. Link to comment Share on other sites More sharing options...
Foxblogger Posted 5 January 2012 Share Posted 5 January 2012 That's some fine work there, Dan. Link to comment Share on other sites More sharing options...
Chairman of the Bored Posted 5 January 2012 Share Posted 5 January 2012 I won't believe it until I see it in 3D. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.