.png)
This is the last of my planned posts on ADLG. It's a postscript to my November post in which I announced that I was taking a break from ADLG for at least a year and follows on from my post about the BHGS ADLG rankings.
Rankings! What are they good for?
In brief, the BHGS rankings are odd. Play in fewer than seven competitions in the last twelve months and your ranking reflects your attendance as much as your performance. Play in seven or more competitions and it is based on your top six scores, dropping your worst scores.1 As I said, odd and very "old school".
The ADLG website maintains a different, international set of rankings based on the ELO system which takes into account the difference in rankings for every game:
The difference between the ratings of the winner and loser determines the total number of points gained or lost after a game. If the higher-rated player wins, only a few rating points (or even a fraction of a rating point) will be taken from the lower-rated player; however, if the lower-rated player scores an upset win, many rating points will be transferred. The lower-rated player will also gain a few points from the higher-rated player in the event of a draw. This means that this rating system is self-correcting. Wikipedia
As such they reflect player performance more accurately that the BHGS version: no dropping poor results.
A comparison
I have always tracked my BHGS ADLG ranking despite never managing more than six events in twelve months. As of 20 October 2025 I was ranked 51st of 182 players up from 64th of 183 in 28 September 2025: a rise of 13 places.
My rise was due to my performance in the 2025 King in the North where I out performed my own expectations to finish seventh.
In contrast, my ELO ranking changed far more as this table shows:
| Date | Games | Ranking (Global) | Ranking (GB) |
Pool (GB) |
ELO |
| 07 Sep 2025 | 37 | 875 | 190 | 332 | 962 |
| 28 Sep 2025 | 40 | 860 | 188 | 332 | 966 |
| 15 Dec 2025 | 45 | 561 | 111 | 338 | 1036 |
The magnitude of the change was probably due to my win against a highly ranked player in the last round of the KITN. Even so the 299 (Global) place improvement is far more than I anticipated. The 77 place improvement in the GB & Ireland subset also surprised me.
More importantly, I finished with a ELO ranking above one thousand for the first time. New players start at one thousand and I quickly moved down before I clawed my way back.
The ELO ranking seems to be doing what it's supposed to. I wish I'd started recording my ELO ranking sooner.
Player pool
Now I want to highlight a major difference between the BHGS and the ELO ranking which has nothing to do with how the rankings are calculated.
The ELO rankings contains nearly twice as many GB & Ireland players as the BHGS rankings.
No doubt this is because the BHGS rankings are temporary. They only list the number of active players in the preceding twelve months. In contrast, the ELO rankings are permanent and life long. I will have one even if I never play again.
This difference is useful as it can be used to estimate the potential scope for growth of the GB & Ireland tournament scene.
Since late 2018 the BHGS rankings have contained between 180 and 190 players. The stability of the figure is quite remarkable. Using the ELO GB & Ireland pool of 338 players, I estimate there are around 148 players who have not attended an ADLG tournament in the last twelve months. I think this is quite high.
This simple calculation is significant because it quantifies the number of inactive ADLG players in GB & Ireland.
An opportunity?
So, there's a decent sized pool of ADLG players who, for some reason, no longer play ADLG competitively.2 The key question is: do they represent an opportunity to increase tournament attendance or it is simply a measure of the normal churn with rules?
The answer depends on why players became inactive in the first place. For instance it could be because of:
- Changes in personal circumstances & work commitments,
- Tournament location & travel difficulties,
- Boredom with ADLG,
- Dissatisfaction with ADLG,
- Dissatisfaction with the ADLG tournament circuit,
- Competing wargaming attractions,
- General disenchantment & players moving on.
The above is a mix of personal and wargaming reasons. The former are difficult to influence, but if it's the latter than something could be done.
Sadly, even using the Internet, I don't think we'll ever know why most no longer compete.3 This makes deciding if there's an opportunity, and formulating a coherent response, difficult.
A solution?
One practical solution would be to improve the rules4 and enhance the level of online support with a view to reducing the barriers to (re)adoption for both casual and occasional players.
For the former, I don't mean a complete rewrite, more the application of polish to aid clarity. For the latter, the current online support is shocking and needs to be more actively managed.5
I've no idea if a new version is in the offing (and don't want to start a rumour), but it's always a possibility. More importantly, I'm not sure what else could be done.
Closing remarks
Estimating the size of any player pool is a perennial problem with wargaming rules of all stripes, so it's good that there's at least one way of estimating it for ADLG.
Whether this represents an opportunity to increase tournament participation will likely remain an open question due to lack of meaningful data.
All the same, a new, more polished, version of ADLG might entice some former tournament players back; if handled carefully.
Finally, if you are an former ADLG player and want to share why you stopped, do leave a comment or get in touch directly.
Footnotes
-
The vast majority of players (81.2% in August 2025) attended fewer than six events. Very few (18.8%) attended seven or more events. ↩
-
Only 78 of 338 players in the ELO database have attended fewer than seven tournaments. Looking at the data again it's likely that 70 of 148 inactive players attended just one tournament. The change arises as I'd confused games with events. The latter isn't in the ADLG data set (as I first thought). Assuming UK events were either 3 or 5 games, one event players must have played less than six games. ↩ -
I listed my own reasons earlier this year. ↩
-
Of course this is always a risk. There are lots of rules that died after releasing a new version: DBMM, FOG to name but two. ↩
-
Before I wrote this post and looked at the ELO data, I uploaded a personal wish list for ADLG version 5. ↩
7 comments :
For an alternative view of the BHGS data see this post by Tim Porter on the ADLG forum.
After Tim's comments, and some confusion of events and games, I have corrected footnote #2.
For those that are interested I prepared a frequency chart from the GB & Ireland ELO data.
Spurred by comments elsewhere, I dusted off my copy of Access and set about analysing the latest BHGS (13 Dec) & ADLG ELO (15 Dec) rankings. Using Access allowed me to analyse, check and cross check what follows very quickly. I used to do this for a living and I'd forgotten how powerful it is.
I had to correct a few entries in the BHGS data not least because I identified a few incorrectly assigned player numbers. The BHGS rankings contained 16 entries for overseas players (according to the ADLG ELO data). There were also four BHGS entries without a player number and who I couldn't find in the ADLG ELO data. This reduced the 191 BHGS records to 171 GB & Ireland players.
In comparison to the figures in the post, I found the latest BHGS & ADLG ELO sets had 171 GB & Ireland records in common. This compares to 338 GB & Ireland records in the latter. So, there's 167 "inactive" players of which 70 have played less then 6 games (so likely to be "one offs"). Or to put it another way: as of mid Dec 2025, there's a pool of 97 "inactive" ADLG players who've previously played in at least two tournaments.
After "refreshing" my Access skills and analysing the BHGS ranking data set from 2025, I found myself well and truly bitten by the data analysis bug. To cut along story short, I've now added data for 2022 (the first full post-Covid year), 2023 and 2024 to make a table of player data covering the four years from 2022-25. Needless to say all sorts of interesting patterns emerged.
As before entries were corrected for spelling errors and missing ELO IDs assigned where possible. Then all overseas players (as defined by the ELO ID) where filtered out along with those without an ELO ID.
[1] The corrected and filtered data set gave a pool 241 players with a valid ELO ID. That's 71% of all the 338 GB & Ireland players with an ELO ID. Of these 60 played in just one year; 40 played in two; 40 played in three and a healthy 101 played in all four years.
[2] I was able to identify players that didn't play the following year. For example 31 played in 2022 but not in 2023. For the following years the corresponding figures were 34 (2023) and 22 (2024).
[3] Likewise, I was able to identify new and returning players each year. For example 2025 saw 25 returning players (played in at least one of 2022-24) and 7 new players (not played in 2022-24). For 2024 the corresponding figure were 12 (returning) and 18 (new).
[4] I also looked at when players stopped competing and if, and when, they returned. 19 players had taken a one year break in 2022-25 and returned. Only one had returned after a two year break.
[5] Finally, I looked at the 160 strong "class of" 2022 to see what happened over the years. 101 (63%) played every year up to 2025, but a number stopped playing every year and did not to return (18 in 2022, 14 in 2023, and 10 in 2024). The balance took a break.
The most important facets to emerge are:
[1] A strong core exists of 101 players who play every year.
[2] There's significant annual churn of 20-40 players. Those taking a break are roughly balanced by new players or those returning after a year off.
[3] A significant number stop playing every year and do not return.
[4] Few players return after taking a break of more than one year.
If you want to see the data behing the above use this link to the full data set
The calculation of the number "inactive" players in the post and the comment above is very rough. It does not take into account players with less than 6 six games (the "one-offs) who played in 2025. Hardly inactive more recent.
When these players are taken into account there are only 55 "one-off" players (all before 2025). This changes the pool of inactive players in 2025 to 114 (an increase from 97 in the initial, rough calculation).
One final note. There's an ongoing issue with duplicates. The BHGS figures treat two pairs as separate players whilst the ADLG ELO rankings do not. It's a pain. It means certain totals differ by two depending on which dataset is chosen as the master.
Then there's the three UK based players with an ADLG ELO record which puts them in a different country. Correcting these increases the pool of GB & Ireland players to 341 or 343 if you use the BHGS dataset. What fun!
This analysis will settle eventually.
Post a Comment