On Equal Score

Forum rules
Please DO NOT POST BUGS on this forum. Please report (and vote) bugs on : https://boardgamearena.com/#!bugs
Post Reply
User avatar
andycupid
Posts: 17
Joined: 31 July 2015, 14:22

On Equal Score

Post by andycupid » 12 September 2020, 09:26

This has been discussed before: the last player to play a piece is the winner of the tie. I've had a discussion with the game designer and he believes that there has to be a way to settle the tie. While this is favorable, I still think the solution is somewhat sub-optimal. As of late, I've lost a handful of games to tiebreakers. What these games have in common is that I played the first piece. The first thing to consider is to make the move as neutral as possible: I don't want X or O to have an advantage because any second player who is decent enough would realize the flaw and choose accordingly. In a game as short as LITS, this is enough to decide a winner, and therefore I'd have to play the first move nearly perfectly. Actually, I'd hope the endgame to be a draw, because that means my opening is flawless.

Now, we know that X has a spatial advantage due to going first (that's the same with almost all abstracts), so I mitigate that by giving O a slight advantage (+0.5~+1), and most players would go second because of that. Somehow going second means playing the final piece more often (I haven't done the math on that...would have to consider board size and piece coverage), and I end up losing quite a few ties. Which is infuriatingly frustrating, because I would spend half an hour to find the best neutral opening just for me and my opponent to play optimally and end in a draw. And I lose. Because my opponent played the last piece. While I applaud players who take me to a draw, I don't think I deserve to lose, and btw, it kind of hurts in a competitive setting such as tournaments or the arena.

In the flip side, how does the player playing the last piece deserve a win?

So I proposed to settle the tie to whoever played the first piece instead. We all know how hard it is to play the first piece, and I find it rewarding to decide the tie in such a way. The designer responded "it breaks the purpose of a pie rule to favor the slicer of the pie", which I don't quite understand, since it's true in other abstracts but rather vague in LITS. At the end of the day, my proposal might be too subjective due to my own experiences (albeit being top level experiences), so I may reconsider an alternative.

I previously played an abstract called blokus, which is quite well known in the abstract community. In both 2-C and C2 variants, the first player has an inherent advantage (+5 and +17 respectively), thus the official tournament games are played in sets (alternating starting player). Some other tournaments use the pie rule, i.e., one plays a neutral opening for the other player to choose going first or second. When there is a tie, it's either a tie, or the game is replayed. I think the game mechanisms are similar to LITS, and therefore propose to modify the rule set to either 1) permit ties, or 2) play over a course of 2 legs, alternating starting player (if still tied, grant win to first player of the first set). Or perhaps, simply granting the win to the first player isn't that bad. Whichever is applied, they're all imo way more preferable than gifting the win to the player who played the last piece.

I very much like the game. It's aesthetically pleasing. Its rules are simple, but it has a lot of depth. It has a bit of chaos in the random openings. But, as I like to say in the forums, the tiebreaker is unhealthy for a competitive environment. Back in the day, this game had a lot more top players, including Daggerheart, who previously posted on this issue and wrote some guides, and many other players who are on par or better than me that I admire. Now they have all gone on haitus, which makes the game slightly less appealing. I may not always be the best, but I strive to, and learn from others to optimize my strategies. Please take the top players feedback into consideration. Else I'd say, I've probably explored all existing strategies, but top level play is ultimately decided like a coin flip.

User avatar
euklid314
Posts: 41
Joined: 06 April 2020, 22:56

Re: On Equal Score

Post by euklid314 » 12 September 2020, 12:59

I understand the arguments (the starting move is in fact the most difficult move) but I think giving the win to the starting player in case of a tie does not resolve the problem of the starting (dis)advantage - it just generates a different "unfairness".

With your proposal the starting player will aim for a tie while the other player knows he has to get one point better.
I cannot judge if then the second player gains too large a disadvantage.

As it is now, both players aim for a win and try to get the last move if it is a close match.

I would be interested in statistics on all BGA games:
How often does the second player change in his first move?
How often does the starting player win/draw/lose (separate statistics with regards of X/O having been changed or not)?
How many of the draws are won by the starting player and how many are won by the player X?

With such numbers one could quantify the starting player inequality and which effect the tiebreaker has right now.

User avatar
andycupid
Posts: 17
Joined: 31 July 2015, 14:22

Re: On Equal Score

Post by andycupid » 13 September 2020, 08:49

That is a very valid point. Hence, I proposed another alternative above. Tbh, a draw is the most accurate result, but replaying a match may get the job done.

Concerning this point, with the current rule set, players can actively search for a tie in mid to end game already. Playing for a draw is not easy, and almost certainly impossible to consciously plan out at the first move. That's because openings are almost certainly flawed in some way. I don't like the idea of playing for a draw, but in reality, it's easier to be done with the current rules than what I suggested.

User avatar
RicardoRix
Posts: 977
Joined: 29 April 2012, 23:43

Re: On Equal Score

Post by RicardoRix » 13 September 2020, 10:43

Come on, be realistic, there is not really an opportunity to play for a draw. No-one does that unless they can see the complete moves to the finish. That's still not playing for a draw just seeing your path to victory given the current rules. It's not like chess, you are mainly trying to optimise each move.

I understand the initial frustration and the proposed solution does sound slightly better than the current, and it's difficult to argue the original ruling because I don't like it much either, but I don't see this as enough of a significant problem to warrant a solution.

I definitely don't like the idea of draw, it's nice to know the game ended with a definitive result.
The other thread: https://boardgamearena.com/forum/viewto ... 16&t=12585

I know you're usually ranked #1 andy, so when you play the other top 5, do a lot of your games really end in equal scores?

User avatar
andycupid
Posts: 17
Joined: 31 July 2015, 14:22

Re: On Equal Score

Post by andycupid » 13 September 2020, 14:28

Regarding the question, no. Draws are infrequent events, it just so happens I most games I lose are draws (it doesn't say so on the result because I would realize the inevitable fate of losing a tiebreaker a couple moves earlier, then try to play weird moves to manipulate the opponent).

True, I'd say casually the deciding factor isn't that important. But if we want to determine a winner in certain competitive occasions, then it would be somewhat important. Now I do lose games here and there, mostly because I played a bad move that got punished, and I rightfully lose. But it doesn't quite as right when you lose because it's a tie and your opponent played the last piece. At least, that means the game was evenly played out, and there is no deserved winner. So, yes, there is no need to modify the existing rules to accommodate such occurrences, but I do like some kind of other ruling in competitive settings. As mentioned, a two leg game is optimal, and I wouldn't even mind the game being played with the same pattern.

When two players play at equal strength, and all the moves are nearly optimal from opening piece to finish, then draws happen. My calculation process every move is to first count 1 move ahead. If my move is worth +4, I'd make sure my opponents can't counter with anything better than that (defensive minded, that is). Then I'll plan ahead and repeat the process. If the game pans out as I planned (sometimes it can be straightforward, sometimes not), then it can very much be a draw. If it doesn't, then my opponent has either made a mistake or outplayed me. Again, draws are infrequent, but still annoying for either player when it happens.

Btw, this game pretty much explains the whole point. One thing to note is that it is indeed a draw; it can be achieved if I played T4 top left. I simply don't feel like ending with a draw, and hence the T4 top right.
https://boardgamearena.com/table?table=111492597

User avatar
RicardoRix
Posts: 977
Joined: 29 April 2012, 23:43

Re: On Equal Score

Post by RicardoRix » 13 September 2020, 21:39

Maybe you have to take solace in the fact, that if a draw result was given and you got 0.5 - then this is the same as getting the win 50% of the time.

On a more serious note, without a better alternative then I think you're a bit stuck.

If you're struggling with the game not giving you anything new (experiences, strategies, etc) then perhaps the best option would be to ask for a bigger grid.

User avatar
andycupid
Posts: 17
Joined: 31 July 2015, 14:22

Re: On Equal Score

Post by andycupid » 14 September 2020, 04:13

A bigger grid...may perhaps make the 5 of per piece a bit more interesting. I don't know if gaining even a single row will exhaust the pieces too easily, but it might deserve a try (at a custom setting, of course).

The 50% win rate is statistically true. Though, as mentioned above, we might want to determine whether playing first does influence the end result.

In terms of enjoyment, the game itself still provides a decent amount. If BGA really were to implement a league ladder system in the future, then I think this definitely needs to be reconsidered.

T72on1
Posts: 60
Joined: 09 October 2019, 12:18

Re: On Equal Score

Post by T72on1 » 15 September 2020, 09:05

I don't mind the draw too much, as I see it an equal part of the rules, and something that can be used strategically. Then again I'm not at the level of andycupid (although not exactly a beginner either), I also almost never play Arena and for some reason tournaments almost never work out well for me :) . So that might make a big difference. Also, it's not because I don't mind that much, that it doesn't mean better solutions are possible. 'Not minding much' is not quite the same as 'it's the best possible solution'.

I agree with euklid314 on the stats. Those might provide us with more insight into the matter. Not sure if it's easy to get those stats though. Unlike andycupid I do aim for a draw if that is the highest obtainable, since I play most of my games in a tournament setting, where difference in points might matter at the end of the tournament iirc.

It would be interesting to see what a bigger grid does. Now indeed it's often only one colour of blocks that is out of stock by the end of the game. I can see it being another strategical variant if we would have to take it into consideration more.

User avatar
gurthbruins
Posts: 12
Joined: 20 August 2017, 16:38

Re: On Equal Score

Post by gurthbruins » 19 September 2020, 13:54

I think 2nd player has big advantage in his option to switch Sides. And now on top of that to win tie is absurd. Better give 1st player the win.
Also, as this is still not enough compensation, always play 2 games, each player having to start once.
But don't spoil the game with other variants. BOL is already fine tuned.
EDIT: On maturer reflection, I have changed my mind. To decide beforehand that first or second player wins ties deadens the game and robs it of the very interesting ways to get the last turn. So I vote don't change the rules, but always play 2 games.

User avatar
motorhead
Posts: 4
Joined: 25 June 2012, 01:37

Re: On Equal Score

Post by motorhead » 23 September 2020, 01:09

Just to add 2 cents from a mediocre player, I totally agree that losing to a tie is an odd call. I kinda think that turning that into a "set of games" in a tournement would help that situation. As in, after a Tie, play to best of two with the other player starting first. Or maybe make all rounds a "Best of Three" where ties count for zero. but going into a round knowing that your opponent might winn off a tie is just an odd feeling.

Post Reply

Return to “Battle of LITS”