Back Alley Oracle #4: Tiebreakers

Tie-breaking systems are an important function of a tournament and today we’re going to jump into the tie-breaking system for Flesh and Blood and dispel some of the myths surrounding it.

What are tiebreakers?

A tiebreaker is a method for determining the ranking of players with the same match points. They are a necessity in any tournament that is not an untimed knock-out format. The swiss-tournament format in particular tries to simulate the results of a round-robin tournament, but without each player having to play every other player. Each round you’re paired up with players with similar results to determine the stronger player, which informs the next round of pairing. In this way, players who win consistently will continue to be paired with other players winning consistently, and in theory, will be playing against provably stronger opponents the more they continue to win. However, during each round and before the top cut, there will be many players who have the same number of match wins, making tiebreakers a necessity to determine standings, a top cut, or an overall winner.

Because a win is just a single match point, it does not describe the magnitude of that win, or how much stronger a player is over another. To address this, tiebreakers use results other than a direct comparison of match points. For example, by estimating the strength of the opponents a player faced based on the record of the player, or the record of the opponents themselves, tiebreakers can provide a reasonable and fair method of ranking players with the same score without having to play more games.

There are many tiebreaker methods used by many different games around the world, and it’s important to acknowledge that there is no agreed-upon “best” tie-breaker method for the swiss-tournament format. That being said, let's dive into FaB tiebreakers.

What tiebreakers does FaB use?

First, the player with the highest number of match points is the higher ranking player.

• Win a match (W) = 1 match point
• Bye (B) = 1 match point
• Draw a match (D) = 0 match points
• Lose a match (L) = 0 match points

After match point totals (MP), the following tiebreakers are used to determine the order of players in the standings (please note that names and acronyms have been updated to increase reader comprehension).

• Cumulative Match Points (CMP): The player whose total match points was higher at the most recent point in the tournament, will have the better tiebreaker.
• Match Loss % (MLP): The player with the fewest losses of played matches will have the better tiebreaker.
• Opponent Match Loss % (OMLP): The player with opponents who have had the fewest average losses of played matches will have the better tiebreaker.
• Opponent Cumulative Match Points (OCMP): The player with opponents who have had the higher average total match wins at the most recent points in the tournament will have the better tiebreaker.
• Randomized: Finally, a random method is used to determine the tiebreaker.

What is the Cumulative Match Points (CMP) tiebreaker?

The primary tiebreaker in official Flesh and Blood tournaments is the cumulative match points (CMP) tiebreaker, which when simplified, can be thought of in the following way:

The player whose total match wins was higher at the most recent point in the tournament, will have the better tiebreaker.

The idea behind this is that, in each round of the tournament, players are paired up against other players with similar records (similar strength). The higher your match points are in any given round, the more difficult your matches will be (because you will be matched against provably stronger opponents). The reverse is also true, the lower your match points are, the less difficult your matches will be.

The cumulative tiebreaker takes this into account by comparing the match points of players in the latest rounds of the tournament where they differ. While the exact process that Flesh and Blood uses is a little more convoluted, it can be understood a little easier using a simple implementation. Imagine each round is assigned a special number. Without going into detail about the math, the most important aspect of these numbers is that they decrease in increasing amounts for each round.

Then, for each round players gain points for each round, and their sum becomes their CMP that you can compare with other players. When you plug in the records of players with the same number of wins, you can see how the record of a player's wins affects a player’s CMP. Here we have 4 players, each with 3 wins and 1 loss, where the loss occurs in a different round for each player.

If you compare Tyler and Nic at the end of round 4, both players have 3 wins total, but when you take a step back at the end of round 3, you can see that Tyler has 3 wins and Nic only has 2 wins. The CMP represents this as Tyler’s CMP is larger than Nic’s.

Another example with a more complex arrangement of 2 wins for each player.

If you compare Nic and Sam at the end of round 4, both players have 2 wins total, but when you take a step back at the end of round 3, you can see that Nic has 2 wins and Sam only has 1 win. The CMP represents this as Nic’s CMP is larger than Sam’s.

In short, the higher your match wins, the stronger your opponents, the more each win (especially your earlier wins) is worth throughout the tournament, and the better your cumulative tiebreaker is.

Are losses better than draws?

No. Absolutely not. The second and third tiebreakers (MLP and OMLP) for FaB take this into account. Let’s take a look at our four players again, this time each with a total match points (MP) of 2 and a cumulative match points (CMP) of 29 (using our example CMP method).

If we compare a draw against a loss in Round 3, Tyler has only 1 loss in 4 rounds (25%) and Nic has only 1 loss in 3 rounds (33%). In comparison, Sam has 2 losses in 4 rounds (50%) and Jo has 2 losses in 3 rounds (67%), making Tyler and Nic have a lower MLP than Sam and Jo, and thus better tiebreakers. In this way, after CMP, draws are better than losses.

If we also look at byes, we can see that they affect this tiebreaker as well. Because Nic and Jo did not play one of their 4 rounds (they got a bye), Nic has only lost 1 of 3 rounds, which gives them a lower MLP than Sam’s 2 in 4 rounds and Jo’s 2 in 3 rounds, and thus a better tiebreaker. In this way, byes are (typically) worse than wins, but better than draws, and losses.

An additional note, if a player loses no rounds, a bye is equivalent to a win because a player’s MLP will be 0%, the same as if they have all wins and no byes. This means that a player that has one or more byes, and does not lose, is not penalized on this tiebreaker.

A loss should never be worth more than a draw. In a timed format, if a loss was worth more than a draw, players can find themselves in a dilemma of who should lose at the end of the game. If a draw is inevitable, both players want a win/lose outcome, but the disadvantaged player has leverage and can refuse to lose purposefully, as the advantaged player would still prefer a lose over a draw for tiebreakers. This leads to very uncomfortable and bad faith situations at the end of games and does not allow the outcome to accurately reflect the relative strengths of the players. For this reason, losses are not worth more than a draw for tiebreakers.

A draw should not contribute towards match points. In a timed format, if draws were worth match points, players would be more heavily incentivized to stall out a game rather than take a loss when it could determine their standing in the MP and the first tiebreaker (CMP). This is especially relevant for Flesh and Blood where tournaments are typically a best-of-one format. For this reason, draws aren’t worth match points, but are instead used as a better tiebreaker over losses for MLP.

How did someone who went X-0-1 beat someone who went X-1-0?

Typically, this is because the cumulative tiebreaker (CMP) is applied before the match loss percentage tiebreaker (MLP).

This is a very common question that comes up when players look only at the total W-D-L, but do not look at when the wins occurred. It is very important to remember that because draws do not award match points - they are not considered for ranking players (MP) or for the first tiebreaker which looks at your record of earning match points (CMP). If both MP and CMP are the same, then it is clear that draws are better than losses when comparing the MLP of two players.

What if I have some criticism about the current tiebreaker system?

As mentioned earlier, it’s important to remember that no tiebreaker system is perfect. There are many pros and cons to every tiebreaker system, and it is dependent on many factors including the priorities of the system itself.

For example, one of the major benefits of the cumulative match points (CMP) tiebreaker includes its resilience to unexpected change. If Tyler has a lower CMP than Nic, Tyler can never pass Nic unless they end a round with more match points than them. Because CMP does not depend on the results of opponents you have played, players can not manipulate the standings of their past opponents by purposefully conceding games. CMP also makes predictions more accurate for top-cuts earlier in the tournament, allowing many players to see if they can make the top cut without having to wait until the end of the tournament when all their opponents’ scores are decided.

That being said, CMP it’s not without its flaws. Losing early games in a tournament can be detrimental to the tiebreaker and can discourage players from continuing to play. It also makes assumptions about the strength of opponents round-to-round without taking into account their performance over the whole tournament, which can skew results with poor matchmaking and/or insufficient rounds in a tournament.

That being said, CMP will continue to be used for the foreseeable future in Flesh and Blood tournaments, but we are constantly revising our processes for improvement and we are not ruling out a change in the future. In the meantime, we recommend the following:

• Ensure that you or your TO are running swiss-tournament rounds with the suggested number of rounds for the players participating. Changing the number of rounds can affect the assumptions CMP makes, especially when there are few players and many rounds.
• Educate others about the cumulative match point tiebreaker. A lot of negativity of any tiebreaker system comes from a misunderstanding of how it works. An explanation does not have to be in-depth or comprehensive, but diplomatically explaining that “earlier wins means stronger opponents and better tie-breakers”, is typically enough to aid a player’s understanding of what CMP is all about.