Here I Ruel – Figuring Out Tiebreakers

Here I Ruel #2

Understanding the tiebreakers

You watch tournament standings and series of numbers you don’t fully understand. You’re wondering if you can play next round or if you should draw to secure prizes.
Today we’re going to walk through tiebreakers in order to make all of those little things a little more clear.

-1st tiebreaker

The first tiebreaker corresponds to the average percentage of matches won by each of your opponents. In order to calculate it, you must add each of their winning percentages and divide it by the number of players you have played against.
However, if someone you have played has a winning total that is less than 1/3, DCI reporter will round it up to 33.33%.
For instance, let’s say round 4 just ended and you played against the following players:
John (current score 0-3)
Bob (1-2)
Billy (2-1)
Yes, I know I’m bad at finding American sounding names, but that’s not the point.
You have John with 33.33% win (rounded up as it’s the minimum score), Bob has 33.33% as well, and Billy has 66.67%. Then your first tiebreaker after three rounds is (33.33+33.33+66.67)/3 = 44.44%

Then, if you have played several times against the same opponent, his winning rate will count once for each time you’ve faced them. This happens only in Grand Prix, Pro Tours and Nationals, but it can still be pretty relevant. For instance, in GP Brighton I played against the same opponent 3 times. When he won his last round it sent me into Top 8 as all of the sudden 3 of my opponents had gone 12-3 instead of 11-4, resulting in them winning 80% of their matches instead of 73% of their matches.

Then as I’m talking about Grand Prix, automatic wins are not the only bonuses which come with awarded byes, as they also make tiebreakers a lot better. Indeed, each bye counts as an opponent with a 100% winning rate.
Therefore, where as the first example with John, Bob and Billy would lead you to a classic 44% and 9 points after three rounds of a GP, someone who would have received three round byes would have a 100%.

-2nd and 3rd tiebreakers

The 2nd tiebreaker indicates the percentage of games you have won throughout the tournament, while the 3rd shows your previous opponent’s games win percentage.
This time there is no minimum rate required for either.

If you win you first round 2-0 your ranking will say:

3 (points)/33.33% / 100% / 0%

Byes aren’t a factor in the second and third breakers as you haven’t played any games in those matches.

Therefore, if you start a GP with three byes and so does your round 4 opponent whom you beat 2 to 1, your ranking on the pairing board will be:
12 / 93.75% / 66.67% / 33.33%

Draws and tiebreakers

When you draw a round, it affects your first breaker the same way as if you had won. While your opponent may have taken 1 point and not 0, he has still played a match and did not win.

The second and third breakers are affected in different ways by draws. For instance, if you choose to draw intentionally, you have to write “ID” in the draw section of the result slip. If you do so, the match will count like no games have been played. Therefore, if you win you first round 2-0 and take an ID in round two, your 2nd tiebreaker will be 100%.

If you don’t have time to finish a match, there are several different possibilities. If you haven’t had time to start game 3 (or even game 2 because game 1 has ended at the very end of the round), then the result will be 1-0 for either player or 1-1. If you were playing a game you haven’t had time to finish though, the result will be entered as 0-0-1, 1-0-1 or 1-1-1. Indeed, any unfinished game counts as a draw.

Actually, it is pretty interesting to see that there should never be a 0-1-1 result at Magic. I’ll give you a minute to try and think by yourself why this is.

Ready for the answer? If you lose 0-1-1, it will count as 0 games won out of 2 in your second breakers, and as 1 game won out of 2 in your third. If you simply concede game two in the last additional turn, your second tiebreaker won’t be affected while your third one will get a little better.
Therefore, you should always concede when you’re in that situation.

At PT Amsterdam my brother had asked if you could ID a game at any point (he was afraid of a boring main deck mirror match against a friend). It was confirmed that could, so they would start from game 2. Out of curiosity, I asked the scorekeeper what would happen if I chose to ID six times in a match (1-0-0-0-0-0-0) in order to get a 0% second breaker. He answered me that DCI Manager won’t accept more than 3 draws in the same match. Not only was the idea stupid, it wouldn’t even work.

The final round

One more round to go, and you have no idea if you can draw into top 8, if a win will make you top 32 or if a loss will keep you out of top 64. That’s when you have to do a little math. In order to do so, let’s take a few examples from the final round of PT Amsterdam.

The first thing you should know is that the last round of a Swiss round tournament is the only one in which the program pairs people not only depending on their number of points, but on their ratings.
Therefore, as long as they haven’t played each other in the same format, 1st seed will play against 2nd, 3rd against 4th, 5th vs 6th etc.

However, if two of them have faced each other already, they will be paired against the next person down the line. For instance, if 3rd and 4th place have faced each other already, the pairings will go like:


I like practicing over theory, so we’ll use a few different examples to see what to do. This will work as a Q&A. I’ll ask a few questions about who should do what and then will give the answers below along with the explanations.

I’ll just make two things clear first. In a Pro Tour, Top 64 get into money and top 50 are qualified for the next PT, the big money is for top 8 competitors.
Usually, you don’t gain or lose more than 2% in the final rounds. Usually, the change doesn’t even exceed 0.5%.

A- As predicted the top 8 in the standings are playing each other. Can Thomas Ma and Kai Budde ID against one another or not, knowing that Kenny Oberg is facing Jan Van Niewenhove?
B- Yann Massicard is paired down in this final round. Does he have any chance to make money?
C- Can I ID my last round against Ludvig Londos?
D- Where will I finish if I win it?
E- Yu Chen Liu and Tobias Wolf want to make top 50 in order to qualify for PT Paris. Do they have to play?

(The answers are below the standings)

1 Nelson, Brad [USA] 39 68.6481%
2 Rietzl, Paul [USA] 38 61.7884%
3 Jacob, Michael [USA] 37 63.3787%
4 Lybaert, Marijn [BEL] 37 59.7831%
5 Wafo-Tapa, Guillaume [FRA] 36 69.5556%
6 Kibler, Brian [USA] 36 67.8201%
7 Ma, Thomas [USA] 36 63.8942%
8 Budde, Kai [DEU] 36 62.6691%
9 Oberg, Kenny [SWE] 34 55.5908%
10 Van Nieuwenhove, Jan [BEL] 33 65.5370%
11 Soh, Terry [MYS] 33 62.0000%
12 Woods, Conley [USA] 33 61.5238%
13 Scott-Vargas, Luis [USA] 33 61.1111%
14 Soorani, Shaheen [USA] 33 59.7749%
15 Cipolleschi, Francesco [ITA] 33 58.3779%
16 Jordan, Dan [USA] 33 57.8120%
17 Postlethwait, William [USA] 33 57.7884%
18 Kawakami, Kazuto [JPN] 33 56.4868%
19 Boeken, Noah [NLD] 33 54.7831%
20 Levy, Raphael [FRA] 33 54.7092%
21 Calcano, Christian [USA] 33 52.8785%
22 Elfgren, Bertil [SWE] 33 50.4422%
23 Cornelissen, Kamiel [NLD] 31 58.5926%
24 Canavesi, Daniele [ITA] 31 55.7963%
25 Pils, Florian [DEU] 31 52.6720%
26 Häggkvist, Sami [FIN] 30 64.8571%
27 Sormani, Reto [CHE] 30 64.4021%
28 Sperling, Matt [USA] 30 62.4683%
29 Brozek, Petr [CZE] 30 62.3704%
30 Carvalho, Marcio [PRT] 30 61.5370%
31 Crisci, Nikolas [DEU] 30 60.3704%
32 Ruess, Jan [DEU] 30 60.3704%
33 Matignon, Guillaume [FRA] 30 60.2540%
34 Rubin, Ben [USA] 30 59.7910%
35 Sharfman, David [USA] 30 59.7778%
36 Caillaba, Germain [FRA] 30 59.5238%
37 van Medevoort, Robert [NLD] 30 59.3333%
38 Brambilla, Davide [ITA] 30 59.3016%
39 Vidugiris, Gaudenis [USA] 30 59.2407%
40 Zhang, Zhiyang [CHN] 30 57.7460%
41 Boggemes, Kyle [USA] 30 57.5712%
42 Watanabe, Yuuya [JPN] 30 56.4127%
43 Rayner, Hugh [AUS] 30 56.0741%
44 Calvetto, Marcello [ITA] 30 55.8291%
45 Nuijten, Julien [NLD] 30 55.2275%
46 Kowal, Brian [USA] 30 55.1905%
47 Nakada, Naoki [JPN] 30 54.8120%
48 Nyström, Per [SWE] 30 54.7090%
49 Elarar, Jasar [CAN] 30 54.5584%
50 Lombardi, Paolo [DEU] 30 54.2621%
51 Lee, Shi Tian [HKG] 30 53.6217%
52 Scheel, Brandon [USA] 30 53.5769%
53 Yurchick, Adam [USA] 30 53.5214%
54 Nakamura, Shuuhei [JPN] 30 52.1667%
55 Osawa, Takuya [JPN] 30 51.0794%
56 Burton, hunter [USA] 30 50.6136%
57 Liu, Yuchen [CHN] 30 48.4444%
58 Wolff, Tobias [DEU] 30 46.2755%
59 Porojan, Raul [DEU] 29 57.0167%
60 Salva, Grahm [USA] 29 54.9103%
61 Görtzen, Simon [DEU] 28 54.5979%
62 Karsten, Frank [NLD] 28 52.4630%
63 Suarez, Andres [USA] 27 63.7118%
64 Hirashima, Yuutarou [JPN] 27 62.4021%
65 Ruel, Olivier [FRA] 27 61.5873%
66 Londos, Ludvig J [SWE] 27 61.3333%
67 Dictus, Mark [BEL] 27 60.9206%
68 Ishii, Taisuke [JPN] 27 60.9021%
69 Craddock, Will [USA] 27 60.2460%
70 Vozdecky, Lukas [CZE] 27 59.7436%
71 Grove, Kevin [NLD] 27 59.5869%
72 Lax, Ari [USA] 27 59.3148%
73 Leitzinger, Gerald [AUT] 27 59.0370%
74 Melis, Bas [NLD] 27 58.8571%
75 Saporito, Thiago [BRA] 27 57.7037%
76 Yasooka, Shouta [JPN] 27 57.5238%
77 Lindroos, Jani [FIN] 27 56.8571%
78 Nakajima, Chikara [JPN] 27 56.6007%
79 Sadeghpour, Johan [SWE] 27 55.9630%
80 Mori, Katsuhiro [JPN] 27 55.8333%
81 Wolf, Christopher [AUT] 27 55.6616%
82 Nassif, Gabriel [FRA] 27 55.2912%
83 Künzler, Matthias [CHE] 27 55.2350%
84 Saowarattithada, Peamw [THA] 27 55.1852%
85 Kurihara, Shingou [JPN] 27 54.4127%
86 O’Connor, Ryan [USA] 27 54.1481%
87 Wallace, Josh [USA] 27 54.0873%
88 Iyanaga, Jun’ya [JPN] 27 54.0238%
89 Unfried, Jörg [DEU] 27 53.7090%
90 Wiegersma, Jelger [NLD] 27 53.7039%
91 Woo, Travis [USA] 27 53.0134%
92 Duke, Ian [USA] 27 52.0085%
93 Schleinzer, Emanuel [AUT] 27 51.8282%
94 Blohon, Lucas [CZE] 27 51.4815%
95 DEAN, PAUL [CAN] 27 51.2090%
96 Lewis, Kyle [USA] 27 51.2088%
97 Ingram, Peter [USA] 27 50.7831%
98 Jacobsen, Kjetil [FRA] 27 49.3571%
99 Holzinger, Thomas [AUT] 27 49.3547%
100 Hatab, Blaine [USA] 27 48.3624%
101 Massicard, Yann [FRA] 25 56.4739%
102 Snijdewind, Ruben [NLD] 24 63.0952%

A-Top 4 players are locked in. Wafo and Kibler can draw peacefully as the worst case scenario would make 9 players 37 points or above. And still, there is no way Kenny would be able to catch that huge difference on them. Then what if Kenny wins and Thomas and Kai choose to draw? Then Kenny will have to catch up a 7.1% delay on Kai which is nearly impossible. Also, as Thomas and Kai’s last opponent finishes with 37 points while Kenny’s would end up with 33, the difference should even get wider.

YES Kai and Thomas should definitely draw.

What actually happened: they drew, and even though Kenny won, he still finished in 9th place, 7.14% from Kai.

B-In order to check for that, you must verify the position in the standings of the person 3 points above him who has the closest tiebreakers below him, which would be 60th place Grahm Salva. In theory, if he won his match Yann would gain approximately half the difference between his place and Grahm’s, so (101-60)/2, so 20 or 21 places, which would still be far from the hoped 64th place. However, if people above him had an interest in drawing that would also need be to taken into consideration when calculating the math.

NO Yann can’t make it, no matter what.

What actually happened: Yann did win and finished in 80th place, 21 place above his previous round’s position.

C-I’m currently in 65th place, pretty close to the money. If I take a draw, I will catch up half the players between me and the person with the closest breakers and 1 more point (61st place Simon Gortzen) but lose half the places between me and the guy with the closest breakers to mine and two points less (101st place Yann Massicard). Therefore I should end up in 65 – 2 + 18 = 81st place.
NO a draw wouldn’t be a very good idea.

What actually happened: I played and lost, ending up in 87th place, when a draw would have made me 78th. I lost two places from a couple of matches who drew. However if those four players had been playing for money, it’s likely one would have conceded in each match, meaning the math would have been more accurate if their results had been more relevant.

D-It is a little more difficult here as I have to assume people with 30 points will ID into top 50. Not all of them, as several of those players must be already qualified for the next PT, but let’s say that out of the 16 matches, half finish in a draw. Therefore, I’ll only be able to pass 6 losers (as several of them have a better breaker than mine), then 3 of the 6 people with one or two points more than me and finish in 56th place. Then of course it isn’t a pinpoint precision, but once the possibility of multiple IDs comes into play, it will be hard to get better.

1What actually happened: I lost but would have finished 57th with those last three points.

E-This is a very interesting situation that Tobias and Yuchen are facing.
When you have good breakers, just like I did, there is not much difference between a loss and a draw.

Indeed, a draw won’t make you pass many players (in my case I would have passed 2 but get passed by 18), but a loss won’t make you lose that many as you’ll stay above most people with 3 points less than you.

On the other hand, the final round of a Swiss tournament is the only time when having a sucky tiebreaker can help. Indeed, a win won’t allow you to catch up with many players who had 3 points more than you, but with half the players between you and them. On the other hand, a draw will make you pass the exact same players, and not get passed by any player with 3 points less but only by half of those with 1 or 2 points less. In Tobias and Yuchen’s case, what will happen if they play? Assuming 7 of the other 30 points match ups end up in a draw and 8 in a play, they’ll get passed 22 players (the drawers and the losers), as well as the two losers of the 31-31 and 31-30 match ups. They’ll pass 24 players, so the winner would finish in 33rd or 34th place, which seems pretty good. However, what happens in the case they draw? Then they will pass all of the losers (8 players) but no drawers, and get passed by the winners of the Porojan-Salva and of the Gortzen-Karsten match. Therefore they would gain about 6 places despite drawing which would be a little short as it would make them 51st and 52nd approximately.

Therefore, in theory NO they shouldn’t draw, even though it is pretty close. Then, if they want to both secure a money finish or if they check once seated at the table that many players around them are actually drawing, YES drawing is the right choice. However, when you draw and you need other players to play, I’d actually recommend you to stay seated at the table and pretend to be playing in order to scare players with breakers slightly better than yours out of drawing.

What actually happened: They drew and finished 49th and 50th! If they had played the winner would have finished 35th and the loser 76th.

I hope this will help you in your future tournaments! Thanks for reading, and have a great week!


30 thoughts on “Here I Ruel – Figuring Out Tiebreakers”

  1. all this did for me personally was reinforce that IDs should be removed from tournament magic outside of top 8 splits and Pro Tours.

  2. I appreciate the attempt to explain this process… but this went over my head. Do people really run these numbers when they decide to draw or play in the event of a tie?! o.O

  3. You are absolutely correct nasty. I Agree 110%

    But that will never happen since players can just choose to play slow, do deliberate dumb things to go to a draw, and it would be nigh impossible to prove otherwise.

  4. Jesse – absolutely. I’ve drawn myself into 9th place twice, largely due to my lack of understanding of this sort of thing. Whenever it comes time to figure out tiebreaker math, I have to someone who’s better at it than I am and ask them if I can draw or not. 😛 So this sort of thing is very useful to me.

  5. You say that %s don’t change by more than 2% which is perfectly valid at the pro tour but if you’re in a tournament with 5-7 rounds you only have 5-7 opponents a couple of whom have probably dropped so if all your opponents still playing lose you’re going to lose quite a bit. I would only bet on that 7% figure being a 90% bet. (of course if you think you have a 90% chance of winning if you don’t ID you’re kidding yourself)
    The big problem with trying to predict this stuff is that people don’t make predictable ID choices and unintentional draws are also possible. In the end for those of you who aren’t good at maths find a friendly maths genius to help at your tournament and make sure you win your first few rounds.

    1 last thing do byes affect your rating or are they just filler waiting for some actual results (which would obviously make your tiebreakers good since your fourth round opponent will start on 9 points so at worst his match wins will be 75% after round 4)

  6. “Shucks, we’ve been playing this whole time and didn’t even manage to finish our first game! Oh, we’re the last match and effectively held up the tournament by 10 minutes? Too bad about that. If only there was this intentional draw thing we could do instead that would have saved all our time.”

    Regardless of whether it’s a good idea or not, there’s no effective way of getting rid of an intentional draw.

  7. Well removing ID is indeed not possible as players will simply play slow etc.

    HOWEVER their should not be a reason that makes you “want” to intentional draw, it should really be punished over an actual result (and if someone is “slow play” a judge will take care of that, while ID is totally free).

    Another point is that the “bye” gives you 100%, that seems quite unfair aswell.

    I mean you “earn” the Bye , but thats allready a free win, that should not also give you another good tiebreaker to “draw” in the Top 8, we really want a Top 8 because the players actual “played” for the victory, and didnt get to that point by draw or simply the Tiebreaker alone.

    I mean if you get a Bye because its an uneven number of players, nobody will blame you for a 100% , but its just different for the Bye’s.

    But no matter what, it won’t change that easy, its that way for so long, and the last one that will have problems with the “fairness” of the system is someone that actual benefits from it // while i blame nobody that “abuses” the system, its how it works, its the job of others to decide about that rulez.

    Just my 2cents :

    PS: Locally we give less prices for ID , like someone at a PreRelease or such, peops just want to see players “play” and not just ID for prices (while the need for an ID is much bigger if its real money).

  8. Can you make an article about the DCI Total points? on how it is calculated? I find it as confusing as tie breakers.

  9. In the first example. John’s current score should be edited to be 1-2, to be consistent with the 44% explanation.

  10. Thank you for the informative article, and your American names were fine (the comment was funny though).

  11. If you need help with some mercan names can I suggest Moon Unit, Dweezil and Diva Thin Muffin Pigeen. They’d be much more authentic.

  12. I’m glad to see this kind of article here – as others have written about in the past, there’s a lot more to succeeding in Magic than just playing well.

    Though that certainly helps 😛

  13. “Therefore, if you start a GP with three byes and so does your round 4 opponent whom you beat 2 to 1, your ranking on the pairing board will be:
    12 / 93.75% / 66.67% / 33.33%”

    This can’t be right. It should be 12 / 75% / 66.67 % / 33.33 %. I don’t get where that 93.75% opponent match win percentage comes. Ok, I do, it’s clearly (100+100+100+75)/4, but that does not make any sense. At that point you have had only one opponent, who has 9 match points out of 12, so that makes your opponent match win 9/12=0.75.

    You definitely don’t get a virtual opponent with 100 % match win when you get a bye! That said, it still is very advantageous to have those 3 byes, because all of your opponents have at least three wins. Compare this to a 4-0 start by playing all the rounds. Only in the best case scenario all four of your opponents will have a record of 3-1, which gives the maximum OMW of 75 %. If you start with 3 byes and win round 4, you’ll automatically have OMW of 75 %, unless you were paired down on round 4.

  14. @pashidar

    Agreed, the OMW calculation for byes seems a bit odd. Against a virtual opponent, OMW should run out like this.

    Round 1 – 0%, he loses to you and goes 0-1.
    Round 2 – 50% he beats his 1st round opponent and loses to you 1-1
    Round 3 – 66.66% he beats his 1st and 2nd round opponent and loses to you 2-1

    End of virtual opponents.

    Even if we assume that the virtual opponents lose only to you and win every other match, the OMW standings if you were to win round 4 should be.

    .75 + .75 + .75 + .75 = .75 OMW. Since it is impossible for an opponent to lose to a player and also have MW % of 100%, it seems that either your math is a bit off or the DCI rules are skewed towards people with byes. I’d lean towards the latter simply because you’ve probably seen such a score sheet.

  15. The part about byes counting as an opp with 100% win is wrong.
    From the tournament rules (http://www.wizards.com/dci/downloads/MTG_MTR_1Oct10_EN.pdf appendix D):

    When a player is assigned a bye for a round, he or she is considered to have won the match 2–0.
    Thus, that player earns 3 match points and 6 game points. A player’s byes are ignored when computing his or her opponents’ match-win and opponents’ game-win percentages.”

  16. “This can’t be right. It should be 12 / 75% / 66.67 % / 33.33 %. I don’t get where that 93.75% opponent match win percentage comes. Ok, I do, it’s clearly (100+100+100+75)/4, but that does not make any sense. At that point you have had only one opponent, who has 9 match points out of 12, so that makes your opponent match win 9/12=0.75.”

    Absolutely correct, my bad! I misunderstood the way a bye would affect your breakers.

    “DCI tournament rules (http://www.wizards.com/dci/downloads/MTG_MTR_1Oct10_EN.pdf appendix D) say game win % is calculated using a game-point system (3 points for a game win, 1 for a draw, 0 for a loss) which invalidates Oliver’s point about conceding 0-2-0 rather than losing 0-1-1.”

    I’ve asked a Level 4 judge to check pretty familiar with those things to check my point was correct and he also though it was.
    In case I have made other mistakes, I can only apologize and recommend you to go check on the link Simon J gave for more details.

    Now concerning the Q&A part, it is true that I’ve made my maths considering a tournament which has about twcie more round than a PTQ. Therefore I should have stated that in the case of a 7 to 9 rounds tournament your tiebreaker will usually be changed by 0 to 1%, but that number could go up to 4.
    However, what is important to know when you’re in the last round of a PTQ is not if you are 100% sure to make it in with a draw, as you rarely are, but to know if your chance to make it is higher than if you would have played. A 50%will be good against a bad match up but insufficient against a good one.
    Let’s say you’re in the following situation before the last round of a 7 rounds PTQ:
    1- 18 pts
    2- 18
    3- 15/69% (1st tiebreaker)
    4- 15/66,2%
    5- 15/66%
    6- You 15/63,8%
    7- 15/63,3%
    8- 15/61,3%
    9- 15/60,8%
    10- 15/59,5%
    11- 13/63,7%
    12- 13/63,5%
    13- 12/67%

    The pairings go as following:

    In order to know who will do what, you should consider first what happens if everyone with 15 points or more draw. Then you’ll end up with 11 players on 16 points or more, and most likely 9 and 10 out, as well as 8 or possibly 7 which is only 0,5% behind. As a result of which 7-8 and 9-10 can’t draw.
    You can now assume that tables 1 and 2 will draw while tables 4, 5 and 6 will play. What should you do?

    If you lose you’re clearly out and if you win you’re in. Meaning playing should give you a 50% chance to make it, regardless of your match up and the level of your opponent.
    If you draw though, you should finish in 8th or 9th.
    First, you must assume you shouldn’t catch up on the five players who were above you before the round. It is not impossible but more than 3% is pretty unlikely. Then, the winners of table 4 and table 5 matches will pass you as well.
    What will determine if you make it into top 8 or not is the result of the 13 points match up.
    You had 63,8% going into last round while they had 63,7 and 63,5%. Those are extremely close totals, but you have played your last round against someone who has finished with 16 points when your contender has played against a 13 points player, meaning that regardless for the results of your 6 other opponents, the gap will widen a little bit.
    It’s pretty hard to say precisely either you’ll make it or not with a draw, but the chances are clearly over 50%, probably about 65%.
    Therefore, even though you are far from sure it will be enough, and unless you know your match up is reaaaaally good, drawing is the right option.

  17. So the pro’s don’t really understand either the system or the maths. Those links to the tournament procedure were really useful for me because I’m quite capable of understanding it and remembering it but working on what people tell you just doesn’t help because most people don’t know.

  18. I drove with some friends to a t2 fnm, prozes were awarded based on swiss standings after 5-6 rounds, going into the last round as the only 2 undefeateds were myself and one of my car mates, so we id’d to guarantee prizes.

    After final standings went up, the judge running the tournament handed out the prize packs, and stated that since we id’s he awarded fewer packs to first and second and redistributed the extra packs to lower finishing players who would have otherwise gotten no packs.


    so … next time we agree to a prize split and play it out i guess since whoever loses is quite likley to get second anyway.

  19. David88 – working things out in your room, with infinite time and nothing on the line is quite different from working them out just after the pairings go up for the last round of the PT (or GP, or PTQ or whatever) and before you have to be seated and start playing. I saw this video of Bucher during the last round of swiss of PT Berlin, he was really confused and trying to do the maths. Then he said he asked someone (it might have been Ruel or Chapin or someone) and they said – you should play. So he did, he won and made Top8. An ID would have given him 9th place. The pros might make a small mistake here and there, but they develop an intuitive sense of how the system works. Hearing a master like Olivier Ruel talk about it is hugely more beneficial than trying to develop the whole calculation in your head, no matter how big it is.

    To all the guys complaining about how good the three byes are – yes, they have to be the best possible result you can get out of your first three rounds. They reward you for being good. They also help the 0-2 scrubs out by eliminating some of the most formidable opponents during the first few rounds. If after 3 rounds, someone with 0 byes is in front of someone with 3 byes, then the system is wrong, because in that case, a pro player could easily have cruised through his first three rounds instead of being “forced” into bad tie breakers.

    To all the guys complaining about IDs – maybe, if you ever were in a position where they were beneficial, you would understand how vital they are. Long tournaments are exhausting and IDs help a lot in the later rounds, especially if you still have 3 rounds to play against the best players/decks of that tournament. An ID is a tool, just like metagaming, connections and good sleeves are a tool. All of the things have to come together for a player to be successful at the tournament, and calculating tie breakers is just one of those things.

  20. ID isn’t a tool, it’s just a necessity since its impossible to stop “intentional unintentional draws”. IDs in the final rounds of a tournament effectively removes one round from the tournament and increase the randomness involved in going to the playoffs. Nothing more than a necessary evil that they have thankfully removed on magic online.

    For swss only tournaments it’s customary to have payouts based on match points rather than ranking. For example in a 5 round tournament, they might give 16 boosters for 15 points, 12 boosters for 13 points and 10 boosters for 12 points. In this scenario it makes sense to let one player win rather than to take an intentional draw. Too bad you didn’t figure this out until after the tournament.

  21. Thanks Oli, that really helped 🙂

    Commentor: Imagine all the matches you’ve ever played were the same format. That’s Total.

  22. Then as I’m talking about Grand Prix, automatic wins are not the only bonuses which come with awarded byes, as they also make tiebreakers a lot better. Indeed, each bye counts as an opponent with a 100% winning rate.
    Therefore, where as the first example with John, Bob and Billy would lead you to a classic 44% and 9 points after three rounds of a GP, someone who would have received three round byes would have a 100%.

    This is wrong, byes do not give 100% OMW. They are just not counted. It used to say 100% if you had only had byes in a tournament, without any real opponent, but wizards recently changed this. (they didn’t give 100% before, it was just “symbolic”)

    Just ctrl + f “Martin Juza” here http://www.wizards.com/Magic/Magazine/Article.aspx?x=mtg/daily/eventcoverage/gppor10/stand3
    And you will see they changed it to state 0%

    This being said, I must add that byes are still very good for your OMW% since most opponents that lose in the first rounds (assuming u won against them, they already atleast have 1 loss) will not win alot of matches. While if you have byes, all your opponent you’r ever gonne play in the GP will have atleast 3 wins. Assuming u don’t get downpaired ofcourse.

    Another misconcetion about OMW% is that player who have dropped from the tournament still get “losses” each round.
    After round 2, you’re 2-0. And you opponent who’m you have just beat drops. His match win % = 50%. This will stay that way for the rest of the tourney.
    You OMW% will offcourse still change since this is the average of the win % of all your opponents.

    I am unsure, but I assume it works that same with multi-day events. So if you go 5-3 on a PT and none of your opponents made day 2 (for example they all went 4-4 so your OMW% = 50%) your tiebreakers don’t get wrecked by that.

  23. Really helpful article. Nice to see you writing for CFB, Olivier. As a long time follower of your limited articles on SCG I hope to be seeing a draft video from you sometime!

  24. Pingback: MTGBattlefield

  25. Awesome article. I’ve always understood TBs but never had the context of “X usually draw” or “tiebreakers shift by Y”. And seeing the question/answer setup like that was a great way to test it.

  26. Pingback: Figuring Out Tiebreakers por Olivier Ruel – Traducción por Inazio Madariaga «

Comments are closed.

Scroll to Top