The FGM Ladder for 2016

[Ithikial]

Hi,

Have not completed one game with an FGM member yet, so, you can take this with a grain of salt!
Agree with the idea that the ladder should encourage participation but not overly reward the most games played - not that this is a bad thing.
Is there some criteria for what constitutes a "ladder" game - QB points min/max, approved scenarios, mirrored games, battle type - probe, etc.
Personally, I prefer scenarios because I'm crap at picking a force.
I have always thought, perhaps mistakenly, that defenders have a harder time than attackers. Should there be a weighting advantage for "defender" winners?
For some spice, there could be a bonus system that awards players who win with, for example, using planes, unusual and/or expensive eqpt.
I'm sure you guys know more about this than me, just a few ideas to throw in the ring.

@Lethal Any CMx2 battle between 2 FGM members is eligible.

The club is generally one step more casual than others, including in the ladder department. There is such a spread of games played and reported already this year as our members have varying levels of RL commitments - the current ladder system reflects that I think. And only Bootie and myself know how close it really is between the players.

 

[Lethal]

Thx for the "heads up."
We seem to have competing interests - the interest of (select ... some, majority, all) players and administrators in having an FGM ladder (and all that means re competition) versus the over-riding principle in providing a fun place for the casual and enthusiast gamer alike.
Unfortunately you will need the wisdom of Solomon trying to find the right balance!
The (in)famous "a camel is a horse designed by a committee" is usually the end result with competing demands such as these.
Let's face it, if you are interested in FGM you enjoy the spirit of competition to some degree - although winning per se may not be the primary driver - the journey of the game is.
So, I suggest if FGM members want to have a ladder, don't be "half-pregnant" about it. Agree on some basic criteria on what constitutes a ladder game so that those interested in participating in a "ladder" event know what the playing field is.
Then the ladder will mean something and members can participate or not, as their whim dictates.
My two bobs worth!
Good luck!

 

[Lethal]

I have been involved with mixed fleet sailing for many years, so am well aware of the challenges of trying to find a rating system for competitors in a non-standard environment. I would be interested to hear from those more experienced in CM ladders the various issues surrounding this. For argument, let's say we wanted the FGM ladder to provide us with a rough ranking of player's participating in a ladder game. We acknowledge this is only a rough rating system and there are many variables that we are not going to address, ie amount of games played per player.

Design a simple CMx2 competition ladder
Task: Record victories/defeats and draws in a format that is easy to administer.

Variables

Game Results (9): total victory/defeat, major victory/defeat, tactical victory/defeat and draw.

Battles Size: tiny, small, medium, large or huge.

Time duration: 30 minutes to 2 hours (15 minute increments)

Standardize
The more factors that are standardised the less a rating system is required, but at the cost of flexibility. I would suggest the following be standardised:

Battle: QB or official scenario
Game Level: Elite
Style: WEGO

That gives players participating in a ladder event the choice of Battle Type (assault, attack, probe, meeting engagement) and Map Size (tiny, small, medium, large, huge).

I think we would all agree there needs to be some reward/encouragement for playing a scenario/QB that is more complex and time consuming. So a weighting will have to be developed with this goal in mind.

For example, ladder points = game result * map size

Game Result: Total Victory = 4 points, major =3, tactical = 2.5, minor = 2, draw =1.
So no penalty for defeat to encourage participation.

Map Size: Tiny 0.25, Small 0.5, Medium 1.0, Large 1.5, Huge 2.0

Total victory on a large map = 4 * 1.5 = 6 points; Tactical victory on a small map = 2.5*0.5=1.25

These figures are illustrative, I would be interested to hear what players think would be suitable.

Administration
I do not know what the administrator's do at the moment, but it does look like a manual system, so a simple spreadsheet could be developed to make the calculations.

Summary
A simple system like this can be gamed. It does require the goodwill of the players. However, it is a reasonable first step in providing a ladder which reflects who are the better players. Refinements could include a player rating and rules that dictate who can challenge who on the ladder - prevent a "ladder watcher" only challenging players below them.

All ratings based ladders are are by definition a compromise of factors.

 

[Lethal]

I have found a free site that should be able to provide a relatively easy method of providing an FGM rating.
Rankade: http:\\rankade.com

I have entered some recent data to see what turns up. If you are interested in seeing what it does please indicate below and I will send you an invite.

The Rhee Weight factor I relate to game size - Medium being normal.

The underlying algorithm requires 5 game results, so it is still a bit flaky. Although called CM:BN I have entered all CMx2 games in this ladder.

It can automatically send emails when results entered and players can enter their own results. I have used the actual members names but have not linked email addresses so there is no way for anyone to link this back to the player.
Other features - charts, etc

Might be worth considering.

 

[Ithikial]

@Lethal

Thanks for your thoughts. Rest assured everything you've stated in your post from yesterday is already taken care of under the current system. Battle size, length and result all matter. The point values assigned to each are a little different but the same intention is all there. The initial switch to this ladder system didn't include averaging out the scores against the games played and it wasn't necessarily rewarding the best players, but the ones that played the most games. (But those that play the most are generally the best I know!) Averaging and hiding the scores on the ladder has encouraged competition and made it a level playing so far for those retired arm chair generals who can have 10 games on the go at once, and those that struggle to get a handful of games finished at any time, juggling work / kids / wife etc. The previous system of not averaging results led to blow outs in scores and it really wasn't a compeitition past June of each year as the leaders were truly untouchable. This time it's different and judging by the marked increase in game reporting by members, it seems to be working.

Also before I took over the ladder about 2.5 years ago, it was built around third party specialty software imbedded with the website. It wasn't pretty and more than once the club lost all it's records to system crashes of forum updates Bootie had to apply. The ladder sits behind the scenes now (in simply Excel) and is safe and secure from any website troubles/crashes. Our members wanted an ongoing record of games and we're trying to do that. Hopefully later this year it will also pay dividends if we can get player 'CM Career Profile' page up and running.

 

[Gnarly]

Administration
I do not know what the administrator's do at the moment, but it does look like a manual system, so a simple spreadsheet could be developed to make the calculations.

I think you'll find there is already a very detailed and automated spreadsheet in place with exactly the sorts of factors and weightings you have described Lethal...



Have a look at the @Ithikial 's (the ladder administrator) latest offering (which is obviously not for ladders) to appreciate his statistical-nerd persona (and I mean that in a good way!):

http://cmmodsiii.greenasjade.net/?p=3847


Edit: GAH! Beaten by the man himself!


Personally, the ladder for me is really just a record of who I've played, and to get an idea of who is a reliable opponent. If I wanted something more serious/competitive, I'd be on the Blitz ladder. But I don't want that, so I am not.

 

[Lethal]

OK, thx for the info.

 

[Lethal]

Hi Ithikial,
First, thanks for your work on the ladder, and the wheels and cogs that make it happen. Appreciate this can be a vexed, although interesting, topic.
Just a layman in these affairs and understand the concept of ranking takes up thousands of hours of brainiacs time around the world.
Is there a layman's explanation of the current FGM ladder/ranking method. Often a lot of the vexation is caused by lack of understanding of the method.
In the sailing world there is plenty of interesting information out there but no-one reads it! Prefer to argue I think.

 

[Cargol]

@Ithikial Still it's quite strange how this ladder works.
Floki which is the ladder leader at the moment didnt report a result for nearly two months and still is sitting at the top of the ladder for at least this much.I think that if the format wont change he will be the ladder leader at the end of the year with 6 games reported in total which is rather unusual if you ask me.

 

[Ithikial]

@Ithikial Still it's quite strange how this ladder works.
Floki which is the ladder leader at the moment didnt report a result for nearly two months and still is sitting at the top of the ladder for at least this much.I think that if the format wont change he will be the ladder leader at the end of the year with 6 games reported in total which is rather unusual if you ask me.

And if I told you he was only there by a solitary point. Less than that if a few more players notch up at least five games.

 

[Cargol]

That wont change the fact that the ladder leader will have as many as 6 games in total in a year

 

[Ithikial]

That wont change the fact that the ladder leader will have as many as 6 games in total in a year

We haven't even finished the fourth month of the year! Still 8 more months of fighting to count. At the current rate, probably another 200 plus games will be completed between club members before the 31st of December.

I chose the minimum five games per year to be counted towards the ladder, based on balance from what we've seen in the past few years. A handful of players (like yourself) are Combat Mission fanatics who play non-stop and can somewhere remember details of more than a handful of games at once. Others not so much. Most players on the ladder do play around 6 - 7 games per year. If we raised the minimum amount of games to be played to be counted towards the ladder to say 10 or 12, then the it would only be a competition for around 6 people by the end of the year. Lowering it below 5 games per year meant it was too easy to cheat the system. Have one or two total victories and then sit back. Getting five straight total victories against other human opponents takes surpreme skill or incredible luck.

There's a graph floating around here somewhere, but most of the FGM memebers also play games that are rated as medium, followed by small and then large. There is no evidence of our members being dishonest and calling every game 'huge' to try and garner bonus points. Which is good to see.

With a few keystrokes I can change the minimum number of games required or any of the points values very easily and quickly. Any changes I make to the database also get automatically carred over historically to previous years for our player database for a consistent multiyear measure. It's a very mallable/editing friendly system.

 

[Cargol]

A handful of players (like yourself) are Combat Mission fanatics who play non-stop and can somewhere remember details of more than a handful of games at once. Others not so much. Most players on the ladder do play around 6 - 7 games per year.
.

Indeed I and a few (good ) others have played quite a few (good) games till now.Unfortunately from September and on I ll be away and my gaming time will be limited so when I ll stop playing I might be having a chance of winning the ladder.
Anyway, keep up the good work.

 

[Ithikial]

View attachment 11635 @Ithikial I really don't understand how Neutral Party (no offense, NP) can be ahead of Cargol in the ladder. NP has played five games, won four and lost one. Cargol has played twenty three games, won nineteen, drawn two, lost two. Could you please explain to us how the new scoring system works?

The Total Points Scored (Based on the Degree of Victory and the Size of the Battle) is averaged over the number of games played. Neutral Party has a higher score than the places 2 and 3 but it's... let's say the closest we've ever had for an FGM ladder. The points aren't displayed to keep everyone on edge. Only way to boost your rankings or maintain your position is to play more games.

 

[Bootie]

@Meat Grinder here is your answer to your question in the other thread.

 

[Meat Grinder]

The Total Points Scored (Based on the Degree of Victory and the Size of the Battle) is averaged over the number of games played. Neutral Party has a higher score than the places 2 and 3 but it's... let's say the closest we've ever had for an FGM ladder. The points aren't displayed to keep everyone on edge. Only way to boost your rankings or maintain your position is to play more games.

By that logic, the easiest way to win the ladder is to play five huge games, achieve total victory in all five, and then never play another ladder game that year. At the end of the year, you would be 5 and 0 and ladder champion.

 

[Ithikial]

True, but what we do know is that virtually all players play different sized battles and never win them all. Winning players 99% of the time the ones to report battle results to the FGM website. There will always be a degree honesty with any type of online competition like this. But if the most important thing for you is to game an online wargaming ladder system... I'd double check why you're playing the game in the first place.

The old system was simply awarding the player that played the most battles rather than looking for degree of victory. The number of battle reports that came in also dropped a bit as players realised they had no chance of catching the leaders from around June onwards. Just trying to correct those faults, and it appears to be working considering the record number of battle reports per month we are currently receiving.

 

[Meat Grinder]

I would like to suggest a possible alternative way of scoring the ladder, a way which rewards players who win (not just play) more games. The winner of a ladder game should receive ladder points for each victory. The points awarded would be based on the level of victory and the size of the battle. The loser should have an equal number of points deducted from their ladder score, or if you want to make losing sting a bit less, perhaps the loser would only have their point total reduced by half of the number of points that the winner received. What this would do is reward players who (a) play more games, and (b) win. A Draw result could be handled by splitting the points evenly between the two players.

Just something to think about.

 

[Ithikial]

Points are awarded as follows. (Same points levels since Bootie first asked me to take over the ladder). The combination of these two are then added to all your other battle results and then divided by the number of battles you've played. I can change the values here and it will auto update the ladder.

 

[A Canadian Cat]

I would like to suggest a possible alternative way of scoring the ladder, ...

No need to reinvent the wheel there is such a system called ELO - https://en.wikipedia.org/wiki/Elo_rating_system. It does not apply perfectly to our type of game but theBlitz uses it on their ladder (note: thBlitz actually has two to scoring systems on their ladder - I personally prefer ELO).

Having said that it is a totally mute point. There was a window of discussion earlier in the year about how to run the ladder and a few ideas were kicked around (including ELO suggested by me) and @Ithikial made his decision on how to score the ladder this year and is now executing to that plan. Next window to discuss how the ladder works I imagine will be towards the end of this year...