*****The OFFICIAL OCUK 2017/2018 Fantasy Football League*****

I have ended up with 4 United Players because Matic moved.

I was going to make a transfer but I HAVE to remove a United player if I want to do a transfer.... Waste of a transfer.
 
Anyone else noticed the peculiar changes to prices? I.e players costing 5.9 or 6.1

Apparently it was the same last year, but I didn't notice it :S
 
The fantasyPL Reddit sub is a goldmine. From what I've read, if 100k people drop a player their price will drop by 0.1. If 200k drop, then it's a 0.2 drop. Price drops are capped at 0.3 though. The same applies to price rises. 100k transfer in, their price rises by 0.1. But not your selling price. That seems to be 0.1 behind. Your players need to rise by 0.2 to get 0.1 more back on your sell price AFAIK.
 
It's not that set in stone price rises/decreases - prices also decrease for players not getting points. But as you note the cap is 0.3 million a week.

The money you get selling a player that has gone up in value is the buy price + 1/2 the price increase rounded down.
 
How does it work then? I've played quite a few seasons and never noticed it or heard about it before. It can't be a very straightfoward algorithm because plenty of players scoring 0pts retain their price for ages.
 
How does it work then? I've played quite a few seasons and never noticed it or heard about it before. It can't be a very straightfoward algorithm because plenty of players scoring 0pts retain their price for ages.
Just keep an eye on the bargain basement players through the season, come end of Jan I'll sign a couple of them to sit on the bench freeing up some more cash to spend elsewhere.
 
Player prices are entirely based on the amount of managers transferring them in vs transferring them out. It doesn't matter how many points they score at all, but of course if a player isn't scoring points he is more likely to be transferred out by managers, therefore making his price more likely to drop.

Hopefully that helps clear it up!
 
It isn't! Wait until the end of the season and see players that have decreased in value below what they were originally worth even though their net transfers are next to nothing - because nobody had bought them in the first place!
 
It isn't! Wait until the end of the season and see players that have decreased in value below what they were originally worth even though their net transfers are next to nothing - because nobody had bought them in the first place!

That's incorrect.

The pricing algorithm also factors in the % ownership as a base, for example Kane (30%+ ownership, 12.5 original value) has seen 164,760 transfers out this week and has not dropped in price, however Shane Long (<1% ownership, 6.0 original value) has seen 7,242 transfers out this week and HAS dropped in price by 0.1 to 5.9. That explains why you have seen very low owned players drop in price despite their net transfers out being relatively low.
 
It isn't! Wait until the end of the season and see players that have decreased in value below what they were originally worth even though their net transfers are next to nothing - because nobody had bought them in the first place!

It must be a very slow decay then, because I've actually said in the past that I found what I believed to be the system (transfer based) to be a bit flawed in that if nobody bought players in preseason there would be nobody to sell them and hence their value doesn't drop! i.e. if a player is overpriced nobody buys them and hence they retain value much better than say a player who had a high ownership at the start of the season and then got sold off. You sometimes see this with 4.0m priced players, people pile in at the start hoping to get value but if they aren't playing they get sold off and drop in value before the 4.5+ players that nobody bothered with.

Looking at examples of players who had extremely low scores last season, some of them retained full value right to the end of the season, with very few dropping more than £0.2m (unless they were players people would actually have bought to begin with, obviously). Now it could be that drops due to not scoring points were offset by players bringing them in but the evidence to me suggests that transfers are far more influential than point scoring. A nobody who scores less than ten points will often drop less in value than a player people have heard of who scored maybe a couple of dozen points but dropped out of the team meaning they got offloaded a lot.
 
Back
Top Bottom