I've just done some calculations based on data recorded within 15 mins of price changes, only looked at players that have changed twice and I get:

Mellberg 0.1 up - 11075 transfers since first price change. Also lowest of the 0.2 risers.

Kanu no change - 11285 net transfers. Confirms your prediction but disproves the theory?

Similarly looking at falls for the only two players that have dropped twice.

I get: Gallas 0.1 down - 13% of those that started with him sold.
0.2 down - 9% of the new number of managers that have Gallas sold or 21% of those that started with him.

Stalteri 0.1 down - 9%.
0.2 down - 10% new or 17% from start.

Tiny sample but clear pattern? Considering margin of error and fact that drops went 6% to 5% to 4% for Gallas and 5% to 4% to 4% for Stalteri.

It was MGP's first price change so it has to be based on B or C.

Subsequent price changes may be based on A, B or C. That's part of their secret and possibly even changed from last year.

C is unknown, so it may introduce quite a large margin of error. You can work the other two out quickly so hopefully it's based on that.


Good call, looking at the figures I put it at 7% with a low MoE because he was in so many teams.

Similar to MGP, Lescott was sold by around 7%, although there's a bigger MoE (+/- 2%) as he was in fewer teams to begin with.

Where did you get this figure from Ferg - both me and Malefice recorded him as +12123 NTI since his first change, which fits with the theory.

I've got Mellberg recorded as -

Cost now 5.1
Cost at last deadline 5
Teams selected by (%) 7
Transfers in 29636
Transfers out 609

and -

Cost now 5.2
Cost at last deadline 5.1
Teams selected by (%) 10
Transfers in 41436
Transfers out 1334

15 minutes after price change (30 at most), could easily be a C&P error but that's where it's from.

Sorry everybody it's my error.

I recall those figures from last night's giving Mellberg a total NTI of 40102, then I used this info from the first night's change:
to get my 12123 result. Clearly my mental arithmetic is not flawless (especially at 2.30am) as that's a thousand out and it should be 11123.

Apologies for that, but it now raises a question because that is less than Kanu who did not change last night.

Are you sure Reds, the figure Malifice quoted is 41102, making the diff 12123?

(or was Malifice quoting your incorrect figure?)

Are there slightly different rules in place for each type of player - goalie, defender, midfielder, forward, Henry?

Not entirely sure, I think I remember it was 40102 but not certain. I expect Malefice did just use my figures, hopefully he will confirm when he next appears.

Just like to give a shout to the guys working this stuff out and say how much i (and i'm sure the others here) appreciate it.

I make it Bridge, Brown, Primus, Bowyer, Campo (currently just over), Ronaldo, Saha, Berbatov to rise next.

Parker, Tonge, Arteta - borderline.

Kanu - mystery.

Nice one ferg. I have been sat here wondering whether to transfer in Brown for Neville and I think your prediction has tipped the balance.

If Berbatov suddenly becomes a casualty, I may be stuck with a dilemma. I had already decided not to bring Henry back in so it's a fairly easy decision to make if one of my defenders is about to lose money.

Am I over-reacting in wanting to not lose 0.1m? I like my players going up 0.1m but I hate them going down.

Will Pedro Mendes fall today?

Hey all, its that time of night again

Do not read on if you are p*ssed, on drugs or are of a nervous disposition.

Please could somebody check this logic for me and let me know if it makes sense...

The formula for price falls is difficult to determine because the % selected by figure is rounded to the nearest full %. For example, if a player shows a % selected by of 1, this could actually be anywhere between 0.50 and 1.49.

Based on 770,000 players this is a range of 3885 to 11578 teams selected by.

That would of course explain the massive variation in NTO % required to cause a fall. Currently anything between 3.5 and 8 % has been identified. Players with a low % selected by (0,1 or 2) suffer more from the rounding because of the high margin of error.

Because we will never know the actual no. of teams selected by we need to determine a system which works for what we do know. I have a suggestion ...

If you subtract 0.5 from the teams selected by % for every player and use that figure to work out the actual no. of teams selected by, this will give you the lowest possible no. of teams selected by and hence will (when x% is applied) indicate the point at which a players value could fall (but wont necessarily fall)

For example...

Ekkotto has a teams selected by of 1%, this actually could be as low as 0.5 % and therefore the minimun no. of teams seected by is 0.5% of 777,000, i.e. 3885.

If we use x% of 3885 as the no. required to initiate a drop, then that is the minimum transfers required to cause a drop.

I plugged this new formula into my spreadsheet and the lowest percentage of any player was 6.24% (so i have asumed the lowest possible is 6%). This is much closer to the figures that have been suggested so far.

I therefore reckon that using this calculation will indicate more accurately when there is a possibility of a players price falling.

This system will not tell you exactly when a players value will fall but it does give an accurate indication of IF it could fall.

This is something we havent been able to establish for price falls yet!


Does this make sense to anyone or am i talking bollox?

Thanks to everyone who has helped with cracking the code in any way. Superb job - you seem to be prettymuch there.

Rises and falls have been updated, Saha up again

I'll plumb all the figures in and see if my theory works !

So for Pedro Mendes

Teams selceted by 5% - 0.5% = 4.5%
4.5% of 770,000 = 34650
6% of 34650 = 2079

therefore 2079 net transfers out are needed for Pedro to fall £0.1mil?

Yes, I used your figures for yesterday.

Yup! Thats what my latest theory is.

However, dont forget that this only indicates that he could fall given that 2079 transfers. The more % selected by the more accurate it is so i would reckon the figure will be close to that (over not under).

If you need the 0.1 to get the right replacement then it's not unreasonable because a 0.1 drop practically means 0.5 at this stage.

If you're just pinching 0.1's in the hope they'll come in handy later then I wouldn't make it a priority if there's something else in your squad that's need fixing.

Current Minimum NTI Required for Price Rise: 11,123 (Mellberg 24/08)

23/08 02:36 BST:

Kenny +14387
Carvalho +13000
Saha +15085 (+46377)

24/08 02:23 BST:

Mellberg +11123 (+40102)
Bridge +18797 (+42538)
Campbell +12363 (+41442)
James +12071
Zamora +14094 (+37253)

25/08 01:55 BST:

Primus +12515
Brown +13689
Bowyer +12984 (+31769)
Campo +12238 (+26627)
Ronaldo +12748 (+25103)
Saha +24967 (+71344)
Berbatov +17259

Under Observation:

Bridge +18560
Kanu +11927

7/8 correct!

Now Bridge has just joined Kanu as the mystery ones.

I did the same thing earlier reg to calculate my MoE, I just didn't show my working

I'm not sure we can assume that they even round though? What if the ranges are 0-0.99, 1-1.99 etc. and it just ticks over when the upper limit is reached?

That's really annoying if my figures are right because I think I've solved the Kanu problem.

He was only added to the game 2 days before it started. When Owen was added mid-season last year his price didn't rise the same as everyone else, he was given a small amnesty period when his price held (presumably to establish a fair user base).

Still worth keeping an eye on Kanu because maybe it works the same way as when a player is returning from injury.

Inexplicable

Who should be moving this morning?

Armstrong, Campbell, Parker, Tonge, Seol, Zamora.

Haven't looked at it too closely and Bridge is still a doubt.

Barton's borderline to drop 0.1 but only know this one because he was in my team.

this thread is making me feel stupid and dumb.