I often get questions through Discord about how Elo is calculated and whether we take faction strength into account.
Do We Account for Faction Strength?
Yes, faction strength is absolutely part of the formula. It’s a bit technical, so the best way to explain it is with a real example.
We’ll look at a recent 5-round tournament that took place after the latest Battlescroll dropped. To spare any potential embarrassment (or inflated egos), I’ve removed player names and the event name.
The winner of the event went 5-0 with Kharadron Overlords, while Soulblight Gravelords and Sylvaneth armies took 2nd and 3rd place.
Step 1: Starting Elo Ratings
First, we check if each player already has Elo data in our system.
Most of the players do, and we use their existing Elo rating. For those who don’t (three in this case), we assign a default starting Elo of 400.0.
Once we have each player’s rating, we calculate the average Elo rating at the event, this gives us a baseline to work with. In this case, that average is 413.6.
Step 2: Faction Bias
Before adjusting player Elo scores, we need to account for how strong their chosen faction is globally and we call this the Faction Bias.
To calculate this, we look up the current global win rate for each faction.
This gives us a positive number for overperforming factions, and a negative number for underperformers.
For example:
A faction with a 60% win rate gets: 400 × log10 (0.6/0.4) = +153.1
A faction with a 40% win rate gets: 400 × log10 (0.4/0.6) = −153.1
Step 3: Elo Formula
Once we have a player’s starting Elo, their faction bias, the average event Elo, and the average faction bias, we plug it all into this formula:
Elo Calculation:
Where K is a dynamic scaling factor:
K = 20 for players with 30+ games
K = 40 for players with fewer than 30 games
xample: Player 1 (Kharadron Overlords)
Let’s walk through the actual calculation for the tournament winner:
| Value | |
|---|---|
| Starting Elo | 586.4 |
| Faction | Kharadron Overlords |
| Faction Win Rate | 57% |
| Faction Bias | +47.55 |
| Games Played | 5 |
| Wins | 5 |
| Avg Elo (Event) | 413.6 |
| Avg Faction Bias (Event) | +1.49 |
| K-Factor | 20 (30+ Games Played) |
Why Use Faction Bias?
Because not all factions are created equal. If you’re going 5–0 with a struggling army, you deserve more Elo than someone doing it with a top-tier powerhouse. Likewise, scraping out two wins with a strong faction shouldn’t earn the same reward as doing it with a low-tier one.
By folding faction win rates into the Elo system, we:
- Reward wins more fairly
- Avoid punishing people for choosing underdog factions
- Keep the system dynamically responsive to the meta
Hopefully, that explains it. But as always, if you have any questions just drop me a message on Discord.
