There have been pieces of thoughtful discussion on other various threads about this topic, and I thought it might be helpful to start a unique thread to focus discussion around how to measure “success” in LOTR LCG.

Opening caveat: some will say that “success” in this game has nothing to do with winning or loosing – it’s all about having a fun and adventurous time. It’s all about theme. It’s all about story. And that is a very fine way to play. In my opinion, that is also a way of evaluating “success.” It’s just not quantifiable – it’s not measurable. Also, by choice, it ignores FFG’s suggestions for how to score games and track results in the quest log. That’s certainly an option. However, for those of us who wish to honor (or are at least intrigued by) the scoring part of the game design and enjoy attempts at measuring (read as quantifying) “success,” here we go…

For the sake of this discussion, let’s imagine you decide to take a week and see how successful you can be while playing a given scenario. (Sound familiar? ) When playing a given scenario, there may be a variety of ways to measure success. Let’s start by exploring two:

1) Lowest Possible Score – no matter how many games you play of the given scenario, no matter how many you win, no matter how many you loose, the bottom line is… how low can you go?

Commentary: This model encourages tracking and scoring wins. Losses become irrelevant (unless you never win). In its extreme, this model by itself inherently encourages deck building with a biased toward calculating the precise combination of cards needed (both in hand and in the encounter deck) that will mathematically result in the lowest possible score. The odds of achieving this low score could be a million to one, but if you play a million times, by the odds, you will eventually obtain that lowest possible score at least one time.

2) Highest Win Ratio – as you play the scenario, you track your number of wins verses your number of losses. The most successful result would be to go undefeated. The least successful result would be to loose every game. Success would be measured in fractions or percentages.

Commentary: This model encourages tracking every single game played – both wins and losses. In its extreme, scoring wins simply becomes a curiosity. When it comes to deck building, the main point (the primary bias) is to obtain a win and avoid defeat in every single attempt. Scoring the win is an interesting afterthought. Surviving every time (or as many times as possible) is the main goal.

Extended Commentary (about Options 1 and 2): In my opinion, I feel uncomfortable with the extreme forms of either of the above options. In the extreme, Option 1 minimizes losses, and Option 2 minimizes the scoring mechanic. Interestingly, I feel that the current FFG scoring system and quest log places premium on Option 1. In all of the formal instruction from FFG, obtaining the lowest score is what seems prized. Period. Feel whatever you like about it, but there has been no serious FFG directive track your win-loss ratio. That said, I think most of us who have been involved in this conversation are not fully satisfied with either of these options (or with FFG’s seeming bias toward Option 1). I think what we are really after is…

3) Hybrid: Highest Win Ratio + Lowest Possible Score – as you play the scenario you log every single game, tracking your wins, losses, and the scores of your victories.

Commentary: This model values wins, losses, and final scores. However, it also requires combining two separate measuring systems and the formation of a new way of ranking player results that factors in both of the measuring systems. Further, there are options on this variant to value score averages and/or your lowest achieved score. Both are interesting, and in some ways land the discussion right back to a renewed version of the debate between Options 1 and 2 from above. Hmmmmm…

And this is where I am left considering. I am hoping that it’s possible to find an elegant way of ranking player results that will factor in both win-loss ratio and final score. I believe it can be done. I would imagine that in the entire history of the world of sport, this kind of problem has been tackled before and met with some reasonable and pleasing result. I’m just not settled on it at the moment, and so I’m hoping that anyone who reads this will be inspired to consider ideas and possibilities and share your thoughts. Perhaps together we can find a more satisfying way.

No one has been 100% happy with the scoring system, and probably never will be. I heartily agree that 'success' should be measured by the fun and enjoyment you had playing it. Losses can also be fun. One other criteria that can be employed is to see who can build a deck(s) that do best against all the scenarios. Again you would have to decide what is 'best'. Probably most % of wins with ties decided on lowest scores for wins. However, this isn't perfect as the longer/higher difficulty scenarios probably score more points. A few scenarios are also dependant on getting the right cards at the right time and can sometimes be won very quickly or take a very long time. It could be argued that a deck that lasts longest under difficult conditions is 'better' than one that wins/loses very quickly.

It all depends on taste. I know that many were complaining that the game was getting too easy - nonsense as there are plenty of ways to add difficulty - so prehaps the measure of 'success' could be winning by the narrowest of margins. ie:- the highest score. Cheers!

I'd measure success first and foremost by the win-loss ratio. To me, that's the main objective when I start a game: to win it. I consider it sorta like a puzzle or any other intellectual challenge.

The scoring system itself is nice to put a number to it, but I'd rather play a deck that shows consistent win % rather than one that fails 75 % of the time, but obtain a great score on the one time it wins.

Great post, I'm in total agreement that a separate thread for this topic makes sense.

Having played in a couple of your tournaments, in order to get the "best" score I find myself taking more and more risks, and hence relying more on luck on the turn of the next encounter card. It is an interesting intellectual challenge to design a deck to get the lowest score, but I don't enjoy the games as much and I don't think it improves my overall game play.

This highlights an important point, whatever scoring system is used, players will end up designing decks and using a play style to get the best score in that system. The scoring system essentially defines what constitutes a good deck and a good play style. In my opinion the current scoring system is driving the wrong behaviors.

The original scoring system promoted decks that took control of the board then healed their heroes and reduced their threat before completing the final phase of the quest. This was pretty tough on Tactics decks which struggle to achieve either sub-goal.

The updated scoring system tried to address this by penalizing decks that "hung around" to achieve these sub-goals. If you weren't able to improve your final score by more than 10 points a turn, it was now better to complete the quest quickly. On the surface this seems like an improvement, but experience has shown that it instead promotes risk taking.

So what criteria would a good scoring system have?

• It would promote player skill (in both deck building and play)
• It would minimize luck
• It would be easy to record
• It would be easy to explain and understand

I'm sure there are others ....

Another thing that makes psychological sense is to make a bigger score better, i.e. a system where a score of 50 is better than a score of 40.
Rather than reinvent the wheel, here are some examples of scoring systems from other sports/games that try to address some of the criteria above.

Reducing the impact of luck.
In sports like Ski jumping and Ice skating there are several judges, each give a score independent of the other judges. The best and worst scores are rejected and the others averaged. Converting this to LotR, we could play 5 games, ignore the best score and the worst score and average the remaining 3.
When it comes to a tournament I would imagine that the order of cards in the encounter deck would need to be set the same for all players, somewhat similar to bridge tournaments. Of course there are cards that cause a reshuffle of the encounter deck, but there still needs to be some luck in the game.

Ease of tracking the score.
Let’s be honest, even though tracking how many turns you’ve played seems like an easy thing to do, it is also very easy to make a mistake and miss a turn or two. Keeping track of time might be easier and just as effective.

Pulling it all together.
You could set a time limit, say 90 minutes, for a competition round. Play as many games as you like in that time. If you’ve played 5 or more games, ignore the best and the worst scores and average the remaining scores to give your result. If you don’t manage 5 games, you don’t get a score.
A loss should score 0 points.
Your score for a game you win should be 75 minus the score worked out using the original scoring system in the rulebook. This ensures a higher score is better. If you want to take time to reduce your threat and heal your characters to get a better score, you can, but you are using up your time!

Example
If you finish on 42 threat with 2 damage on heroes and 6 victory points your score would have been 38. Your new score is 75-38 = 37 points.
If you play 6 games with scores of 37, 0, 40, 32, 60, 0 you would ignore one of the 0’s and the 60 and average the others (37, 0, 40, 32) giving a final score of 27¼

i agree wih comments on low scores. they involve luck, risk taking and doing it dozens of times. you might have the lowest score in the world, but how many tries did it take you? a few? a hundred?. . my friend and i have a point margin we gauge our decks on.take in account the difficulty of the level, your win/loss ratio to see how successful we were with our decks. if some one could come up with a formula that factored win/loss ratio, score averages and difficulty, you would come up with a satisfactory way to gauge yourselves.

but bottom line for us --fun. to mix it up and challenge ourselves, we'll randomly choose heroes and build our decks accordingly. we'll play a scenario untill we beat it a couple times. after that it's for challenge and fun, not a lucky high score.

this is lotr lcg. not chess. billy fisher was the best that ever played. he spent every waking moment studying the game and analyzing it. how much fun do you think he had? the best chess players usually become paranoid mental cases. some people play to be the best that ever was. others play it for fun. to challenge themselves is sufficient. if you're a mathmatical genius, yeah, you'll probably bore easily with an lcg. you should play chess.

I consider win % the best indicator of success. That being said, if there was a way to assign a "score" to a failed game then you could take the average score of all games played against a specific quest as the score. Would have to come up with some way to assign a fairly high score to a failure - perhaps based on the difficulty level (i.e. a higher difficulty level equates to a higher score for a loss). The score for losing a scenario needs to be significantly higher than what the score for a good win would generally be (i.e. don't want losses to end up scoring better then wins except in some extreme cases).

#1 metric of success for me - Am I having fun? I do roughly track Wins and Losses on the different scenarios with different decks though.

RGun said:

What numer do you think that would be? If you did say, 25, with a rating 7 quest, a loss would be 175. Does that seem fair? I think it could be.

Another approach that would include % of wins into the equation would be to use a weighting factoring based on the % of games won. The formula could be (AVERAGE SCORE OF GAMES WON) * (TOTAL GAMES PLAYED / GAMES WON). So for example, if you played 5 games and won 3 with scores of 75, 100 and 125 then your net score would be: ((75 + 100 + 125) / 3) * (5 / 3) = 100 * 1.67 = 167. If you had won all 5 games for a 100% win percentage then the weighting multipler would have been 1 and your net score would be 100. For a tournament, everyone could be required to play the same number of games - e.g. play a scenario 5 times. This would avoid someone getting lucky and winning the first 2 games and then stopping. Pick enough games so that the luck factor of lucky draws from the encounter deck gets nullified to some extent.

You could also divide by the quest difficulty rating if you wanted to have a comparison between scenarios. e.g. ((AVERAGE SCORE OF GAMES WON) * (TOTAL GAMES PLAYED / GAMES WON)) / QUEST DIFFICULTY. Although ratings don't seem to be an exact science so not sure how useful this would be.

Both win ratio and scores should be combined to measure a player's success. If you only use the score, then you'll see those rushing games like in the recent tournaments, which is also fun, but doesn't say much about the player's ability and the quality of his deck. If we'd only use the win ratio though we 'would be back at where we started - players satlling their progress deliberatly until they have enough allies and attachments for an easy win. I dislike the latter which is why I prefer the new scoring system which has been a big step into the right direction.

For a tournament I could see the following measuring system:

You play X games and add the points of your wins. Divide the result through the win ratio.

Example:

Player A plays 10 games. He won 8 of them and scored 100 points in each.

8 x 100 = 800

win ratio = 0.8

800 / 0.8 = 1000

Player B plays 10 games. He won 5 of them and scored 60 points in each.

6 x 100 = 600

win ratio = 0.5

600 / 0.5 = 1200

Player C playes 10 games. He won all of them and scored 110 points in each.

10 x 110 = 1100

win ratio = 1.0

1100 / 1.0 = 1100

In the example, player A would win the contest as he had better results on average than player C and was more successful in total than player B. Player B would loose against player C despite having the far better results, because his win ratio sucks.

Edit: crossposted with RGun who apparently has the same idea

There are lots of good ideas so far, I hope we can start to pull some of the better ones together into a coherent scoring system.

I've had a few more thoughts on how to measure success. I started with the question.

If 2 players compete against the same quest, both play 5 games and both win 5 games, which is the better deck?

What if they both play 10 times and win 10 times?

What if one of those decks manages to control the threat level, always winning on zero threat. Does that really make it better than the other deck?

If the zero threat deck wins 9 out of 10 (with zero threat) and the other deck wins 10 out of 10 (with an average final threat level of 45), which deck is better?

Of course, there is no right or wrong answer to these questions. The thing is, asking those questions helped me realise that maybe we don't need to consider threat at all. Maybe all we need to count is the number of VP's acquired and subtract points for each hero that died. Lets say 5 points per hero death. If you fail a quest you are considered to have lost all your heroes. The higher the score, the better. Interestingly this means if you play a 2 hero deck then you only subtract 10 points when you lose which may open up some more options.

Play 5 games and add up your scores and you have a pretty good measure of how consistent a deck/player is.

This method has a few key advantages.

1. It doesn't penalise you for not playing spirit to reduce your threat. Threat still plays an important part of the game (engagement checks/secrecy/losing at 50) but it is not part of your final score
2. It doesn't penalise you for not playing lore. You don't need to heal your heroes to increase your score.
3. You can play Tactics as its meant to be played, putting damage on Gimli is meant to be a benefit after all.
4. It is really simple to track. Add up your VP's at the end of the game and take off 5 for each dead hero. Done

Any thoughts on this idea? Can we really ignore threat and damage and end up with a good measure of success? I think so!

Memetix said:

Okay, this would definitely solve a part of the problem. The other issue, namely the rpoblem of those Speedy Gonzalez rushing decks could be solved by using the win/loss ratio.

Perhaps Zjb could implement both in his next tournament with as many layers participating as possible to get a broad basis of feedback.

leptokurt said:

I think my proposal already accounts for that, not as a direct ratio but by assigning a negative score to losses.

Memetix said:

Yeah, right. I also think that your idea of rejecting the best and the worst score would make sense. Don't like the thought about a fixed timelimit though.

RGun said:

Okay, I really like this!

And here's why:

1) It completely respects the FFG scoring system, not altering it in the least. That feels quite important to me, actually.

2) It could work well in a tournament setting in that it offers a straightforward way of valuing both low score and win percentage over a set number of plays.

Very nice RGun! x1,000

leptokurt, your idea is basically the same too, just arranging the formula in a slightly different way. I think I like RGun's arrangement the best because visually it factors the elements of score and ratio in a very clean way... (score)*(ratio) = result

Of course, I'm eager to consider other ideas around this discussion over time too, but for me personally this feels quite satisfying. Satisfying enough to try this out for at least a tournament or two in February and see how it actually plays. Exciting!!

Really, the only drawback I feel is that it forces the same number of plays for everyone participating in the tournament. Not that huge of a deal, but I'm sure no matter what number is selected, some people will wish for more chances and some will wish for fewer. It limits the flexibility there - but maybe appropriately so.

Other random thought:

I've seen several people make comments (on a variety of threads) that amount to people saying that it does not take skill to build a highly risky speed deck. I disagree with this statement. I think it does take skill. It just takes a very specific kind of skill that also minimizes the value of other possible skills that could be incorporated into game play. It does take smarts to figure out the speediest combo. It's also (in my opinion), over time, less satisfying of a game play experience. But it's not dumb or unskillful. That I feel strongly about.

Thanks for the feedback juicebox. I think it's important to find the right balance between how much lows scores versus win % contribute to the overall net score. Don't want it weighted too heavily in either direction (i.e. don't want win % to totally dominate so that it almost makes low scores irrelevant). I didn't have a chance to assess my formula too much to see how much of an influence win % has - after we see some results it could be tweeked sightly to add a modifier multiplied against the win % to increase/decrease the weighting that this has on the overall net score to find the right balance.

I also think in a tournament everyone should have to play the same number of games to minimize the luck factor of someone getting good encounter draws in the first couple of games and then stopping. However, my formula uses the average score so it doesn't really require everyone to play the same number of games in order to get valid scores for comparison purposes...a variant could be everyone has to play a minimum of x games, but with no cap on max games you play. This eliminates the case where someone gets 1 good (but lucky) score and then stops.

RGun said:

That's a nice variant idea. I agree that a minimum number of plays prevents someone from posting one lucky first attempt and stopping. I'll think about this more, and maybe in February I'll run one tournament without the variant and one tournament with the variant, seeking feedback as we go. I may also try and do a poll before the February tournaments to elicit feedback ahead of time.

Oh, and I forgot to say another reason why I like your idea:

3) Losses have impact in this system. Losses matter - in life and in games, both.

juicebox said:

The Rabbit decks might take some skill in building the deck, but not in piloting it. If you make a deck that sees less than 10% of the cards IN the deck, and a similar number for the encounter deck? Where is the skill? Where is the decision making .. there is none.. you are just slamming cards and powering though. You are completely relaying on luck, luck in your starting hand, luck in your encounter draw and luck in your card draw.

Is it a legit tactic? Sure. I think the "risk" vs "reward" factor in the rules and scoring system is pretty cool. Like you can build and run such risky decks and get awesome scores. That doesn't mean the Rabbit deck is any less luck based and THAT is the crux of it.. luck has nothing to do with skill.

The great thing about the Rabbit decks is that they are very risky, and it is a big tactical decision to run one in a comp, as you know that MOST times you play it you may loose to a single treachery card or drawing a big monster. So do I think it is a legit deck.. sure I do. Do i think it takes less skill to run one? Yes. Do I think the score system should reflect the losses so Rabbit run decks are not the ONLY legit deck to run? Absolutely

RGun said:

Dude, we are talking about a game that children play. My 9 year old son can not do this math. It is confusing to even adults. I mean look at your instructions!! You are going to introduce a new problem, input error and do not say your not. You have fractions, % calculations, rounding, and averages. .. . . what a mess. You also do not say what you do with your remainders... as it is not 1.67 it is in fact 166.6666666666667, so witch one will people use and even if you did say that you have placed a "rounding" factor on top of it...

Still the REAL problem with it, apart form it being so unwieldy, is that your score system encourages conceding. Like why would I play to the end of a match when I can just concede for a score of 0 that has nearly no effect on the final tally. At least not the same effect as limping over the finish with a high score. The ENTIRE point of trying to do this score system is to minimise people grinding at a quest and just playing it a billion times until they get a good score. Your system doesn't really address that, when you can just call concede get a score of zero and have nearly no effect on your final score. In fact your score system encourages conceding, witch was what we are trying to eliminate in the first place.

We got to just keep this system SIMPLE.

On the plus side, if we are all playing the same number of games (say 5) then the formula simplifies to

"Add up all your winning scores, divide by how many games you won then divide by how many games you won again."

That seems pretty simple.

booored said:

We are talking about a game where you have to be a lawyer to understand all the rules and you complain about some simple maths? Using a win/loss ratio without having periods is almost impossible, as a ratio is per definitionem a quotient. If one can make sure that the denomintaor is always a 5 or a 10, this problem could be solved though.

I know what you want to say about the problem about conceding, but I think that's part of the general scoring system in which an epic fight against a series of mean encounters gives you a lesser score than rushing through some easy encounters. However, as some folks already said: the more you loose, the bigger the impact. Even a good deck and a good player can loose from time to time, if they draw some bad encounters, so I don't consider it a big problem if a single loss has only little influence on the overall result, as it allows you to take at least some risks instead of paying it safe all the time.

the point though is that you can concede at anytime.. like say yo play and you are going to win but you do not like your score.. you can just concede and get a better "meta" score by calculating zero.

Example....

Using the values in the 1st post...
((75 + 100 + 125) / 3) * (5 / 3) = 166.6666666666667 (1.67)

BUT.. what if I did the same.. but decided to concede two of them at the last minuet... this calculation becomes...

((0 + 100 + 0) / 3) * (5 / 3) = 55.55555555555556 (55.6)

so basically what this score system dose is say... just concede ANYTHING that is not a good score. In fact playing a ton of games at 0 is better than just getting a great score 1st turn. This system encourages conceding even more than the old one.

So yeah.. this is useless even worse than the original, good idea .. but needs work to iron out these problems

oh wait.. hang on.. I think I have this all backwards.. the calculation would be...

((0 + 0 + 100 + 0 + 0) / 1) * (5 / 1) = 500

Hmm.. ok .. I see how this works now... still complex.. but your right it dose work a lot better than I thought.

Yeah, I'm astonished, too! I thought I could trick the formula somehow. But that's not possible. At least as long as the score you want to take out of calculation is less than the double of the average of your other scores. Basically, this is because we're dividing 2 times through the number of the games won. (The exact value is not 2 but 2.25).

Example1 : 4 games won with score 100, 1 bad game won with 220.

Without counting the last game as a loss: (100 + 100 + 100 + 100 + 220)/5 * (5/5) = 124. With counting it as a loss: (100 + 100 + 100 + 100 + 0)/4 * (5/4) = 125. So you're better of not counting it as a loss.

But example 2: 4 games won with 100, but the bad game has been won with 230:

With the bad game: (100 + 100 + 100 + 100 + 230)/5 * 5/5 = 126. Without the bad game: (100 + 100 + 100 + 100 + 0)/4 * (5/4) = 125. So here it would be better to willingly get all heroes killed and lose the game.

How often does it happen that you have a bad game that has a score more than 2.25 compared to your other scores? I guess it could be a few times, because of Dead Marshes (consider a game where you have to skim the encounter deck twice for Gollum...)

We could even improve the formula if we divided 3 times through the number of the games won. So the new formula would be:

(sum of all scores) * (g * g) / (w * w * w)

where w is the number of the games won, and g is the number of all games (won or lost or conceded). The factor needed to take advantage of willingly losing a game is (3g² -3g + 1)/(g² -2g +1). For 5 games, this is 3.8125. Examples like above:

Example 1: 4 games won with 100, 1 bad game with 380.

With the bad game: (100 + 100 + 100 + 100 + 380) * 5 * 5 / (5 * 5 * 5) = 156. Without the bad game: (100 + 100 + 100 + 100 + 0) * 5 * 5 /(4 * 4 * 4) = 156.25

Example 2: 4 games won with 100, 1 bad game with 390:

With the bad game: (100 + 100 + 100 + 100 + 390) * 5 * 5 / (5 * 5 * 5) = 158. Without: (100 + 100 + 100 + 100 + 0) * 5 * 5 /(4 * 4 * 4) = 156.25

How many times do you think it happens that someone loses a game with more than a factor of 3.8 of his points than usual?

And what do you think of the new formula in general?

So, today I played 5 games of the tournament for this weekend, just to double check how hard it was and if it was winnable. I went 2-3. That means the weighted factor is 2.5 times the average score, ouch! So the quest is winnable, but scores could be high if you lose more games than you win!

Pete... go back to your physics thesis!