F1 Fantasy Tools
patreon
The potential, new Price Change Algorithm for 2025
🕑 Added 2025-03-14 14:24:12 +0000 UTCThis post will go over everything we currently know about the potential, new Price Change Algorithm for the 2025 F1 Fantasy Season.
Let's start off with a disclaimer: all of the info below is based on dummy data left in the game by the developers at launch, so none of this is verified by real data. This means that it’s possible for the actual algorithm to be slightly or even completely different come Sunday. Use this info at your own risk!
1. On what data is this analysis based and how did we get it?
When the game launched, there was data for the first 3 races of 2025 present in the game - presumably to test the site before launching. The devs tried to hide it, but not very well since it’s visible on https://fantasy.formula1.com/en/statistics right now.
The data consists of points, prices and price changes for all 30 assets over 3 races. 90 data points in total.
2. Price Changes per Price Tier
There are 2 price tiers in 2025:
A Tier = above 20M
B Tier = under 20M
Each tier has 4 performance categories: Great, Good, Poor & Terrible and symmetrical price changes for those categories as shown in the image.
3. The Algorithm
At its simplest, the algorithm is: “Calculate the average points per million (AvgPPM) per asset and see in what performance category it falls.”
AvgPPM = the sum of the 3 latest scores of an asset divided by 3 ; divided by the current price of the asset.
AvgPPM thresholds: 0.2, 0.3 and 0.4.
These form ranges for the different Performance Categories shown in the image.
4. The Issues
1) Race 1 and 2 are different
Because we always divide the sum of the latest 3 scores by 3, and the first race (and similarly the second) doesn’t have any previous scores to use in this sum, we basically take the average of (0 + 0 + currentScore).
You can look at it as if there have already been 2 races before Race 1 where every asset scored 0 points.
This means that the AvgPPM values will (on average) be significantly lower for the first 2 races than those from race 3 onwards BUT the AvgPPM thresholds of 0.2, 0.3 and 0.4 stay the same.
As a result, price change direction will tend to be to the downside for the first 2 races.
2) The AvgPPM thresholds are too low
The second, and in our opinion the most glaring issue is that the AvgPPM thresholds are too low. Last year’s average PPM for drivers was 1.0. Now to get a great performance, you only need an average of 0.4. This will cause serious inflation from race 3 onwards.
Luckily, it’s also the easiest part of the algorithm to tweak, so we think it’s possible these thresholds will still change.
5. The Implications
IF this is the way it’ll work on Sunday, the best way to avoid the early possible price drops, is by having as few B Tier assets as possible, since those are hit with -0.6M instead of only -0.3M for A Tier assets.
From race 3 onwards, the reverse is true: B Tier assets will increase faster in budget than A Tier assets.
Another implication is that because everything is solely based on PPM, it’s more important than ever to find the underpriced assets who should get higher PPM values and therefore better price changes.
6. Some Visualizations
In the scatterplots of the first image (also available in higher quality attached to this post), all 90 available datapoints are plotted by AvgPPM and price according to the algorithm described above. All 90 price changes are correctly explained by the algorithm into the different performance sections.
The second image is the same, except more zoomed in on the thresholds to better see some of the more interesting data points.
At the bottom is a legend of all the data shown in the labels.
7. Some Credits
A lot of detective work by a lot of different amazing people in our Discord:
JW
Mikael
DanSmithKY
Jim2828
LeapsFrog
Freakz
ZaelChronicles
And of course rhter!
You can join all of the discussions here: https://discord.com/invite/uvUfg3VNXk
8. Some Closing Remarks
Price changes in the Team Calculator (https://f1fantasytools.com/team-calculator) are based on this algorithm, so use them with care.
Our personal opinion on the algorithm:
Fundamentally, we think this will be the algorithm for 2025. However, we also think it’s possible they’ll make some adjustments to the specifics after a few weeks or maybe even for this Sunday. The 0.2, 0.3, 0.4 thresholds are nice round values that are likely easily adjustable to something more fitting like 0.4, 0.6 and 0.8.
But whatever adjustments (if any) they end up making and how they’ll affect the optimal strategy for budget building is anyone’s guess at this moment!
Comments
Alex B
This is fantastic stuff! Pints for the Team 🍻
Delermain
Amazing work!
Roy van der Meer
This is great! Thank all of you!
