The first round is always determined by randomness. By excluding it from the SoS counting, you would exclude some elements of random.
the more the tournament advance in rounds, the more the ranking gets refined and definitive. The last round you played should have more importance than who was your random opponent in round 1, to tie-break who is your next opponent.
if you’re a top player in that tournament, that no-round-1-SoS would only rely on top games. If you manage to stay up, then you should be compared to your same level opponents, excluding that (maybe yes / maybe not) talented player you had in round 1.
if you’re a bottom player, that SoS would only rely on down games. If you stay down, then you should be compared with downers, not using the first random opponent you had. That would exclude the luck some of your same-level opponents had or had not.
first round SoS have no use whatsoever to determine round 2 pairing.
So why do we even count the first round in SoS ?
In a 4 round tournament, I’m even considering this as a better solution :
SoS = 0x opponent 1 + 1x opponent 2 + 2x opponent 3 + 3x opponent 4.
(but this could be (0x, 1x, 1x, 1x), (x0.5, x1, x1, x1.5), or (0x, 1x, 1x, 2x), or anything else).
By round 4, you’re playing people with a similar number of wins to yourself, there’s no need to put extra weight there, it’ll already be a high value if you’re at the top tables, or a low value if you’re nearer the bottom.
Your round 1 opponent is indeed random, but you can’t discount the value of that round. Say I have to play the National Champion in round 1, and get beaten - at least I’ve got a good SoS out of it. If I’ve have been paired against someone easier to defeat I may have actually won instead of losing. That SoS value represents the difficulty of the match played - it gets more accurate as games go on and there’s more data and more opponents, but the inherent worth of keeping track of relative strength of opponents is appropriate at all levels.
By your system if it comes to tie breakers, we ignore someone playing the National Champion in round 1, that didn’t count for anything, but someone who got him round 2 (after saying beating someone who’s never played in a tournament before in the first round) gets full credit. How’s that reasonable?
Also, even if doing well, if you’re paired against good players early, you can’t play them later, so you’re then giving more points to players who don’t encounter other top ranking players until later in the rounds (and penalising people who had to play them early).
The Swiss system means a good player will be getting better and better SoS as they progress, as they are also paired against other people who are winning most of their games. Giving additional artificial weight to this seem unnecessary and with no clear benefit.
I played in a tournament last weekend, 5 rounds Swiss, no cut. At the end I was tied for first with another player with 14 points each. We played each other and there were 3 other players we both played against. The only difference were round 1 opponents - mine finished at 6 points and his at 10, so I ended up 2nd on SoS and IMO this was more fair than deciding by a coin flip. Of course I had no control over who our first round opponents were, but still his win was harder to get.
Disregarding the first round SoS is not right, but if you wanted to smooth the SoS, you could drop the worst and best score. To be honest you are losing information either way and I wouldn’t recommend it.
In my opinion, if pairing are random, results should be influenced by that random factor.
eg. high skill player vs a random noob => the high skill player should win, because he have more skill and was paired with a random lesser skill player => hence that result of that 1st round is more due to a random factor than an actual well-made, acurate and fair pairing. And those happens in last rounds.
Well, assuming the national champion will go high ranking in that tournament, someone who get to play him in round 2 allready won his round 1 ? So that round 2 opponent have more legitimity than the round 1 lesser skill opponent the nat champ had ?
In round 3, players get to play better level opponents, or worse. That last round you made vs players of your skill is more important, to me, than the game you made 2 rounds ago vs players nobody could say they were of your skill or not.
And the more you get rounds, the more precise everybody know if you were doing well in a tournament or not, the less random pairing are and the more fair they get.
So the more important last round should also weight in the SoS than the previous, to me.
Because pairing are less and less random the more you get rounds.
Hence that progression of weights.
Granted, you lost vs the local netrunner champ in low rounds. But what is important to show your real strength to break a tie with players of your skill is how you played vs players known of your skill, not of those of skill too far from you that everybody knows you must win or loose.
It’s not focused on smoothing, because to me smoothing does not help to break ties.
It’s “legitimazing” the SoS.
For a 4 round tournament, if I choosed (0,1,2,3), counting 6 scores, at the end I have more dispersion in players’ SoS than a regular (1,1,1,1) counting 4 scores.
If I choosed (0,1,1,1), counting 3 scores, then I have less dispersion than the actual 4 scores SoS (and so, more smoothing).
If really no-round-1 is a problem (it is not, to me, because round 1 is throw-a-dice-get-your-points, but I understand…), we could talk only about (0.5, 1, 1.5, 2), (1,2,3,4) or a fibonacci suite (1, 1, 2, 3, 5, 8, 13, etc).
I still don’t really understand this mentality as a tie breaker. By that logic, shouldn’t using a bye tank your SoS, since you have made the first round substantially easier? (Note: I am not actually advocating this, but just trying to follow the logic of “harder to get x prestige” = “more deserving of win.”) We also don’t settle ties by giving preference to the professor and next design, even though players with those IDs certainly had a harder time accruing prestige than they might otherwise have had…
No. You’ve no way of proving that. As stated in my original response, the person playing in round 2 could have played a complete scrub in round 1 or even got a bye. How is their game against a high value opponent in round 2 worth more than anyone else who played him/her in round 1? Both people would be playing against the same tough opponent.
By devaluing the first round, you’re artificially skewing the data in an unfair way.
Did you fail to get into the top cut of a tournament recently? It feels like you’re trying to solve a problem that doesn’t exist, or at least replace something you view as flawed with something else that’s just flawed in a different way.
Actually, he organized a tournament (he finished last btw) and he was frustrated because third and fourth players were tied in prestige and SoS.
Yes his idea has flaws but it’s genuinely well intentioned.
@MrMint Why do you say I was frustrated by that ? They were 4th & 5th, and I explained there that since the pairing were based on regular SoS, then the “new” ranking had flaws ? I did not even put that in the same thread there, not to interfere with results ?
(having 4 points in a tournament with Apex / Chronos doesn’t mean anything in the current Turing meta btw).
I’m sorry but end of round 1 you have :
Top player of all sorts : 4pts
Second Top player : 4pts
Third Top player / lucky guy : 4pts
4th Top player / lucky guy : 4pts
Unlucky man 1 : 0
Unlucky man 2 : 0
Scrub 1 : 0
Scrub 2 : 0
That category :
Top player of all sorts : 4pts
Second Top player : 4pts
Third Top player / lucky guy : 4pts
4th Top player / lucky guy : 4pts
It seems to me it’s slightly more accurate and fair than the round 1 7-random-guys-category to pick an opponent for mister #1 at the end of the day.
And it will get more accurate and more accurate the time being. So, more valuable, to me.
If you take the case “a good player, let’s call him player 2, fights the top player, player 1 and fail by an inch in round 1”, then if player 2 win all other rounds (since he’s a good player), player 2 may have to tie break with another good player (player 3)
The player 3 say, lost only one match vs player 1 in round 3.
Player 3 will end being ranked 2, yes, because he fighted good players the whole tournament, and player 2 who had to climb back, was fighting lesser players.
You count your first round opponent SoS because he was an opponent that you faced, and SoS tells us how difficult of an opponent it was.
It doesn’t matter that the pairing was random; in fact, because it’s random, the first round SoS is very important. Two players won their first round, which one’s higher ranked? The one that played a tougher opponent.
Byes don’t affect SoS. When counting SoS, you ignore rounds that had a bye. I think that’s actually an advantage for the person who got the bye, if for no other reason than it’s the only way you get a 1-1 record and 100% SoS. But if you’re getting a bye, it’s fairly likely that SoS won’t end up mattering for you. (Unless it’s an Earned bye. In which case, go you!)
If you replace all byes with a player named Larry, who is so bad at Netrunner that no one can ever lose to him ever, and he played all the people who would have gotten byes, the tournament produces the same information about the skill of the players at the end. Yet in the current system, if you get a 32nd person to sign up on the day of the tournament and he is an awful Larry, the people who get free wins off him get punished in SoS. But if Larry stays home sick, the byes don’t effect SoS and we somehow assume that someone who fought B+ players for 5 rounds is worse than someone who fought A- players for 4 rounds. Doesn’t make sense to me.
Would you rather play an awesome player that you have a 40% chance of winning a game against in round one or one that you have an 80% chance of winning a game against? Prestige is still more important than any amount of tiebreakers. The luckiest matchup you can possibly have in round one is one you easily can win, not the one that’s likely to give you the best tiebreakers, (because it’s not a given that you do win).
Sure, you can call the guy lucky AFTER he beat the best player in round one, but when round 1 started I bet he wasnt super happy about his pairings. That’s the whole point of the tiebreaker: it tries to estimate how hard it was for you to win your matches, on average, compared to other people. If you have lower tiebreakers, it’s because you played against easier opponents. What difference does it make in that regard whether you played the best player in round one or round six?
In this system, weigths could be (0,1,2,3,4,5) or (1,1,2,3,5,8) or (0.5, 1, 1.5, 2, 2.5, 3) or anything else.
I did not found a magic formula yet, of course. I think you guys convinced me that removing round 1 is not a good idea though.
That’s what I did, I called him lucky after the match. I do get your point.
But the player you describe still was random paired vs a better player. Both managed to win in the exemple but the second one was paired to a guy who made low ranking.
That involves luck, to me. A little more than matches in round 2 and later rounds, where you’re paired with “guys like you”.
Both player were talented enough to win their matches. The best out of them is still the one who had the luck to be paired with a guy “just under his skill” (or with a better matchup or opponent with agenda fload or etc).
Whether that match was hard or not, the prize is still here. You won prestige points, will still have better SoS than second player from this game and “next game, please”.
If round 1 pairing was though then after round 2 your round 1 opponent should still have better Prestige than the second player’s opponent. So your SoS should have a little bump over the second player.
Round 1’s pairing & opponent doesn’t reflect at all your skill and involves the more the “luck factor”. Starting rounds are the most “feared” by top players for a reason (I know you said you don’t care, but let’s say you’re a “special” player A lot of top french player I know don’t like that moment.). That “fear” is to not to get stuck in the “creative field” (where I sit usually btw).
Round 2 is a little more accurate evaluation of your skill thus removing some luck factor
Round 3 is a little more accurate and remove some luck factor
Round 4 is a little more accurate and remove some luck factor
Round 5 is a little more accurate and remove some luck factor
Round 6, let’s say final round, is the best evaluation of your skill in this tournament.
Weither you made good ranking or bad, final round involves the best “guy like you” opponent in the whole tournament <=> the luck factor in that final round’s pairing should be of the least importance vs your actual skill / non-skill reflected by Prestige allready.
By giving increasing weigths, to me, the luck factor importance in SoS is nerfed and the skill factor importance in SoS is boosted.
Using a tie-breaker in, say, round 4 pairing, could use a SoS with (0,1, 2) weigths or (1,1, 2) or (0.5,1, 1.5) with the 3 first non-magical weight exemples.
That sort of SoS could reflect better I think where you & your next opponent are in the competion of skill rather than giving the same importance to the luck factor in round 1 and to the little less luck factor in round 2 and 3 with the regular system which weights (1, 1, 1).
It is more fair, truer to skill level, to me. The luck factor get smoothed in late rounds and the skill factor is sharper. So to me, it’s a better tiebreaker.
It all seems kind of ridiculous to me. If two players have exactly the same opponents they had the same difficulty level and should have the same tiebreakers, regardless of the order they faced those opponents. You’re creating way more variance by weighting the rounds. If I play against the best player in round 1 and it’s barely worth anything for tiebreakers, I am going to be fucked on all fronts.
@anon34370798 I think you kind of miss the fundamental point of SoS ratings by complaining about the “randomness” of it (i.e. the lack of ability of the player to affect their pairings).
SoS is not supposed to measure anything about the player themself. It is supposed to measure the difficulty of the “puzzle” that they were set (i.e. the opponents they have to play over the course of the day).
If you have an issue with some players getting easier “puzzles” than others for whatever reason, that is an argument for changing the rules for how the tournament system matches opponents (to presumably try and ensure all players face similar SoS over the day), not for changing the way you measure the “puzzle difficulty” (SoS).
Turning the question around a bit…would weighting help the case when you have Unlucky Top-Rated Competitor who gets paired against Scubbity McScrubface* at random in round 1? If you are a player with a legitimate shot at making a tournament cut, how badly does it hurt to get randomly paired against someone who goes 0-fer on the day and has a negligible SoS?
*This asked by someone at the Scrubbity McScrubface level.