PL predictions

[Making Plans 936]

http://www.dailymail.co.uk/sport/football/article-2385951/Chelsea-tipped-win-Premier-League-Bloomberg-forecasts.htmlSeems a pretty fair assessment even tough I think it understates our potential by at least 10 points.   [img]http://i.dailymail.co.uk/i/pix/2013/08/07/article-2385951-1B2FB97D000005DC-995_634x411.jpg[/img]

[thegali 6]

Newcastle 8th? They forgot to add an extra column "the kinnear factor"

[I am a Banana 0]

Stoke tipped to finish below us, but have a better chance at top 4 ?!

[ricardo 7,339]

[quote user="Making Plans"]http://www.dailymail.co.uk/sport/football/article-2385951/Chelsea-tipped-win-Premier-League-Bloomberg-forecasts.htmlSeems a pretty fair assessment even tough I think it understates our potential by at least 10 points.   [img]http://i.dailymail.co.uk/i/pix/2013/08/07/article-2385951-1B2FB97D000005DC-995_634x411.jpg[/img][/quote]

LOL, Newcastle 8th Wet Sham 10th.  Bloomberg are financial experts and on this evidence you can see why they didn''t see the Economic Crisis coming.

[Vazzza 62]

I appreciate this is a computer generated result by using previous stats and lots of fancy algorithms and stuff, but I still call shenanigans on something that at least can''t round up/down the points total so everyone doesn''t have a stupid point tally with decimal points. Whats the thinking behind that, trying to look fancy??

[Rufus Smalls 0]

This thing has massively underestimated the likelihood of drawn games - it only predicts 3 teams to draw 10 games, whereas in the previous 3 seasons the number of teams to have drawn 10 or more matches are 13, 11 and 14. I guess the effect of predicting fewer matches as draws is that the simulation doesn''t predict the usual massive bunch-up in mid table, instead the teams are fairly evenly spaced, points-wise. Over the course of 38 games I suppose there''s more chance of stretching the teams out if more game are predicted to end in 3 total points, as opposed to 2 total points for a drawn game which will keep the points totals closer together. In any case, I think they are on the money with Chelsea for the title, and I also agree that the new boys + Stoke are looking good for the drop. Would have thought we would get more than 41/42 points though...

[AJ 0]

Newcastle to finish 8th?. What fucking idiot came up with that nonsense?. They''ll be lucky to stay up

[Cerberus 0]

As with any forecast model, it all depends upon the robustness of the assumptions and the impact of randomness (i.e. luck, injuries etc).However, of the three predictions (Fink Tank, Bloomberg and Statto) I have seen so far, there is a fair degree of clustering.

Team	Fink Tank	Statto	Bloomberg	Average
Man United	70.8	82	80.3	77.7
Man City	74.9	74	80.8	76.6
Chelsea	73.3	71	81.7	75.3
Arsenal	73.5	72	72.8	72.8
Tottenham	72.4	70	66.8	69.7
Liverpool	69	61	64.8	64.9
Everton	61.7	63	54.3	59.7
Newcastle	50.8	46	47.6	48.1
Southampton	49.5	44	45.9	46.5
Aston Villa	47.5	46	44.9	46.1
Swansea	44.5	46	46.9	45.8
West Ham	43.4	46	46.8	45.4
West Brom	45.8	46	43.9	45.2
Fulham	43.7	46	43.9	44.5
Norwich	41.9	46	41.5	43.1
Stoke	41.7	44	39.8	41.8
Sunderland	38.7	43	42.9	41.5
Cardiff	38.8	37	39.2	38.3
Hull	37.3	31	33.5	33.9
Crystal Palace	32.9	28	32.1	31.0

[Making Plans 936]

Well you can argue the toss over Newcastle''s & West Ham''s predicted final position or the prediction that Stoke have a 0.1% better chance than us of making the Top 4 - which we all know is never gonna happen for either Club this season.For me, the interesting thing is that in this prediction there''s only 6 points difference between 16th position (42 points) & 8th position (48 points) and I think that''s a pretty accurate stab at the final outcome.  That being the case, if we can beat last seasons tally of 44 points by just 6 points, and I see no reason why we shouldn''t, then we could be looking at 8th position.