Ah! So this week we will see maybe a shake-up in the make-up of the tournaments?

Yes, the new algorithm went into effect for this week. Most people probably won't notice a difference because it was okay for them in the first place. A few outliers may see slightly different competition. As always there's a random element so you will on occasion be placed higher or lower than you should.

I noticed. The gaps between players' KO points are enormous now.

I know it's just getting started and we'll see how it plays out over time, but in my tournament, at least, GOAT is not going to stand for Greatest Of All Time but Sacrificial.

I enjoy playing KO. I've got 509 KO points with 53 wins. If I didn't like challenges, I still wouldn't be playing either KO or FT after 9 years, and over 9mil points. I have no issues with the way KO was set up (before the recent tweaks). I realize that it will get tougher as I move into the higher brackets. currently, I've been playing among the divisions #21 through #15. I always agreed with the adage, "If you don't play against tougher competition, you won't improve." I, personally, like challenges, obviously.

I'm seeing tougher competition, and getting wiped out in the first week. How about everyone else? Are you doing better, worse, or the same?

So far, I'm seeing a lot of familiar names with the same results - good days, bad days.

I started playing again a couple of weeks ago - still got wiped out but at least I didn't get pulled into Set 1 as my reward, which is what used to happen.

It's working better for me. I'm not permanently in Set 1 anymore. And I've avoided the rabbit, too.

So what are you going to do ? Duck the rabbit?!

She's more likely to deck the rabbit - don't ask me how I know!

There is a player in featured tournament 73 with 407 KO points and 63 weekly wins this week. That seems a little unusual doesn't it? Most of the other players in that bracket have around 100 KO points and some have yet to even win a weekly KO! Seems like a mismatch....

Hey Nas, I think that might be called the 'fox in the henhouse' gambit!

Absolutely!

I was just coming to post this but you beat me to it.

Terry has some splainin' to do!

Is there a reason that player was placed into that tournament? I'm more curious than anything else but I do feel kinda bad for the other players in that tourney.

Perhaps Whodat's the cat among the pigeons? Not his/her fault for being in that round, but rather a "fowl" play for the others in the tourney.

Any ideas then?

The mentioned player not only won tourney 73 but comfortably had the four highest scores in that tourney bracket. It goes without saying that the player carries no blame for being there, we don't choose where the system puts us!

I need to read back about the changes. I wasnt aware they had occured as I have stepped away from Knockout a bit. Even in the "old way" you would see cases like that every now and then with a player dropping well below their expected placement range for tourney numbers. In those cases affecting my tourney I have been both the fox in the henhouse and one of the 15 chickens being brutalized


I always assumed it worked something along the lines of a random selector that drew players from the top 48 or so names one at a time to fill brackets, starting with #1 and working down the list. So someone among the top 48 in the registered pool of players would be expected to pop up somewhere in the top 3 tourneys but there is a chance they could slip lower, with decreasing likelyhood as you move further and further from their natural range..and that pattern would persist all the way from top to bottom.


It has to be something along those lines because the "out of bounds" players always seem to be lower than expected rather than higher.

It's a long way down though Answerman.

The player in question has WAY more weekly wins than anyone featured even in tourney 31 and a similar number of KO points.

Can you tell approximately where they should have been? I think the biggest one I have seen was roughly a 35-40 tourney drop..

that was me. I was in the very bottom tourney after I had been a knockout regular for at least a year or two. I felt ridiculous up aganst a bunch of new players with 0-5 wins.

then I got worried I would run into the next generation's superstar in their debut and suffer embarassing defeat to a false underdog...but no. It was as you describe here, lots of mismatches and an easy weekly win.

That felt worse to me than when it was the other way around and there was some heavyweight beating up on me and/or my peer group.

Terry made it sound like he was adjusting the coding for it to make this less likely or impossible. I can get behind that as long as variation is maintained and we dont see too much of the xact same group from week to week.

It's as Nasty Liar reports, namrewsna, Whodat has roughly the same number of KO points as those in #31. There is only one other player in Tourney 73 with more than 100 KO points (142) and the rest average considerably below 100. Lowest man on the totem pole had 72. Must have wondered what hit him!

I don't think the changes affected the chance of dropping down really far. Only the ranking criteria were changed. As far as I know, selection works as follows:

1) Rank all the players
2) Put the top X of them into a pool (X might be 16, but is probably more)
3) Randomly pick one player from the pool, put that player into tournament 1 slot 1
4) Fill up the free pool slot with the next player in ranking order
5) Randomly pick a player for the next tournament slot
6) Repeat until everyone is assigned.

The change would have affected only step 1.

The chance to "survive" (drop down) one set of X drawings is roughly 1/e = 36.79%. If X were 16, the chance to drop 10 tournaments would be a bit better than 1 in 100,000. 20 tournaments, 1 in 10 billion. 30 tournaments... probably not in this universe . If X is larger, the drop would be appropriately further - so for X=32, you could expect to drop down 20 tournaments with 1 in 100k plays. X=50, the 1 per 100k chance would give roughly 31 tournaments.

By the way, we could roughly determine X by observation if we knew the exact ranking formula and my selection algorithm conjecture is correct. Tournament 1 slot 1 must always be a player from the top X, so if we, for example, find the #42 ranked player in that slot, we'd know X is at least 42. If we were to find a player ranked 70 in tournament 2 slot 3 (19th drawn), we'd know X is at least 70-19+1 = 52.

I am curious to know the true algorithm and randomness that decides how players are ranked because it must be so real variable randomness for that to happen.