How long does form last? {THREAD}

I’ve been looking at some stats from last two seasons. My goal is to get some insight into how many weeks different players should be owned before transferring them out.

#FPL #FPLCommunity
The idea behind it is that we would ideally jump between the players with the best form while timely drop those players that have seen their good form end.

My goal is to try to figure out approximately how many gameweeks a player’s form typically lasts.
In a similar vein, I want to know how many weeks of patience for a return to form is optimal for a player currently performing below average.

I will now describe the method I use. In a hurry? Skip to the analysis!
METHOD🔭

Step 1: I look at a player’s seasonal average stats in goals and assists.

Step 2: I break his season into chunks of gameweeks and compare each isolated chunk to his baseline season average. I just experimented with dividing these chunks into 4, 6 and 8 gameweeks.
Step 3: I begin by looking at the 4 first GW's (so 1-4) and compare these with his season baseline. I then do the same for GW 5-8 and so on. When all that has been done I can see if it matters, for example, if I owned Mané in chunks of 4 weeks versus chunks of 6 or 8 gameweeks.
Step 4: I come up with a final number of how many times owning the player in that type of timeframe ended in a positive or negative result.
So if Mané in GW 1-4 overperformed against his baseline season score I add a point in the 4 week category, but will subtract one point if he were to underperform in GW 5-8, and so on.

If the time period is very similar to his baseline season stats then I don’t add or subtract.
PROBLEMS 😕

I’ve been looking manually at these stats and one disadvantage is that I look at gameweeks 1-4 and 5-8 but I realize that a more accurate method would be to look at every 4 week period which would be 1-4, 2-5, 3-6 and so on.
I felt this would be very time consuming so I decided against going there, but it's something that I would recommend doing for the players that you consider for your team.

So let’s start the analysis and go through each player that I have applied this method to...
ANALYSIS 🧐

Mané:

4 weeks: Difference = +2

6 Weeks: Difference = -1

8 Weeks: Difference = -2
Conclusion: Holding on to Mané for 6-8 weeks seems on the surface to be a usually bad idea. This is a player you really want to punt short term when his fixtures turn, or just get him in as soon as he hits his form because it might not be long until he dips.
Salah:

4 Weeks: Difference = +5

6 Weeks: Difference = +1

8 Weeks: Difference = +1

Conclusion: Salah seems more reliable than Mané medium term, but also much streakier in short bursts.
Sterling:

4 Weeks: Difference = +2

6 Weeks: Difference = -3

8 Weeks: Difference = -1

Conclusion: Sterling seems to be streaky and owning him for the right 4 gameweeks looks ideal.
Aubameyang:

4 Weeks: Difference = -2

6 Weeks: Difference = +2

8 Weeks: Difference = +1

Conclusion: Looks like it’s better to hold Auba and be patient with him instead of trying to hit his form in a short window of time.
KDB 17/18+19/20 seasons:

4 Weeks: Difference = 0

6 Weeks: Difference = +2

8 Weeks: Difference = +4

Conclusion: The pattern is clear, be patient with this guy! KDB, Auba and Salah are the ones with plus in ranges of 6 & 8 GW's which support the idea of them being consistent.
So what do you guys think? Is this a good way to predict and utilize the length and timing of a player’s form? Is the data perhaps too random or can we glean some knowledge from it?
You can follow @fpl_gents.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.