Patch 16.14, measured
Protoplasm Harness is the one change on patch 16.14 that moved by more than luck
38,832 recorded games on patch 16.14, counted up to . 6 items changed in this patch and every one of them was measured.
stamp filling recorded games 38,832 items measured 6 moved 1 too close 5 not enough games 0 window either side 10 days recomputed
Every item changed in patch 16.14, measured
Change in wins per 100 games: games containing the item after the boundary against the same read before it. Every figure is Measured. The champion chips show who finishes the item on patch 16.14, the patch the item pages are built for. 108,616 ranked games from EUW, Korea and NA, found by following Diamond players and the people they play with.
Protoplasm Harness
games after the patch against the same games before it
-1.5 wins per 100 games(range -2.9 to -0.1) 21,570 games Measured 7,011 games on the thinner side of the boundary, 10 days each way. Games before the patch beat games after it.
Navori Flickerblade
games after the patch against the same games before it
+0.7 wins per 100 games(range -0.6 to 2.1) 23,914 games Measured the range covers zero, so this could go either way 7,998 games on the thinner side of the boundary, 10 days each way. The range covers zero, so this could have gone either way.
Hextech Rocketbelt
games after the patch against the same games before it
+0.38 wins per 100 games(range -0.33 to 1.09) 85,554 games Measured the range covers zero, so this could go either way 28,390 games on the thinner side of the boundary, 10 days each way. The range covers zero, so this could have gone either way.
Rapid Firecannon
games after the patch against the same games before it
+0.3 wins per 100 games(range -1.0 to 1.6) 24,308 games Measured the range covers zero, so this could go either way 8,050 games on the thinner side of the boundary, 10 days each way. The range covers zero, so this could have gone either way.
Phantom Dancer
games after the patch against the same games before it
-0.3 wins per 100 games(range -1.6 to 1.0) 27,492 games Measured the range covers zero, so this could go either way 8,637 games on the thinner side of the boundary, 10 days each way. The range covers zero, so this could have gone either way.
Kraken Slayer
games after the patch against the same games before it
-0.23 wins per 100 games(range -1.13 to 0.66) 52,094 games Measured the range covers zero, so this could go either way 19,062 games on the thinner side of the boundary, 10 days each way. The range covers zero, so this could have gone either way.
Games with Protoplasm Harness in them went down after patch 16.14 landed, measured against the same games before it. Its whole range sits clear of zero, so the change showed up in real games rather than in the notes only.
The champions of patch 16.1410 of 484
The highest scored champion and role pairs measured on this patch. Win impact is games won per 100 played, against the average pick in the same role on the same patch, and every figure is Pooled: part the pair's own record, part the role average, mixed by its games. The full ranking, with every band and every floor, is the tier list.
Syndra Mid · 3,793 games
A 
Thresh Support · 4,640 games
A 
Twisted Fate Mid · 2,496 games
A 
Bel'Veth Jungle · 1,230 games
A 
Poppy Support · 1,519 games
A 
Rell Support · 2,134 games
A 
Locke Mid · 3,240 games
A 
Wukong Jungle · 2,410 games
A 
Xerath Bot · 2,701 games
A 
Ahri Mid · 3,755 games
B A pair needs 30 recorded games before it is scored. The whole tier list for patch 16.14
These numbers come from 108,616 ranked games we recorded on EUW, Korea and NA, found by following Diamond players and the people they play with. That is not every rank and it is not every region. Everything on this page is patch 16.14. Counted up to .
How to read the report cards
Three verdicts, and each one means something different about the cards above.
This is not the same measurement as an item win rate. A win rate counts finished builds at the end of the game and quietly drops everybody who bought the item and lost early. A before-and-after read drops nobody, because the same kind of player sits on both sides of the boundary.
1 of 6 items moved by more than their own range. Those are the cards whose verdict reads beats: the whole 95% range sits clear of zero, one way or the other.
5 items came out too close to call. Each of those ranges covers zero, so we cannot say which way the item went. That is not a broken measurement. It means the change was smaller than what 38,832 recorded games can see, and a site that printed a direction there would be printing the middle of a range as if it were a finding.
Every item changed in this patch had enough recorded games on both sides of the boundary to read. Nothing on this patch is waiting on more games.
A report resolves when both sides of the boundary carry enough games. When one side is thin the report still prints, with its wide range and the word not enough games, because the thin side is a fact about our crawl and not about the item.
Read the range before the middle number. A change of half a win per 100 games with a range from minus one to plus two has not been measured, it has been guessed at, and the range is how you can tell.
How this patch was measured
The short version. The cards above say which part of this they came from.
Riot ships a patch on a known date. We take the recorded games in a window either side of that date, split them by whether the game contained the changed item, and count wins. The difference between the two sides is the effect, and it is printed in wins per 100 games because that is a unit a player can act on.
The trick is that the two sides are otherwise the same. Same queue, same rank band, same regions, largely the same players, and only days apart. That is what a before-and-after read buys you over a win rate: nobody is quietly dropped from one side.
What the two game counts mean
The games inside each estimate are every game in both windows added together. The thinner side line under it is the smaller of the two windows on its own. They can be far apart: a report can stand on twelve thousand games in total with eighteen on the thin side, and it is the eighteen that decides whether the report resolves.
When either side falls short, the report prints its wide range and says not enough games. It is not removed and it is not rounded into looking certain.
What this page does not measure
Champion changes have no boundary report here. The before-and-after pipeline runs on items, so a champion buff or nerf has no card yet, and inventing one from a win rate swing would be exactly the mistake this page exists to avoid. The champions section above is the tier list's own measurement on this patch, which is a ranking, not a before-and-after read.
Riot's own patch notes are not reproduced here either. This page says what our games did after the patch, which is a different question from what the notes said would happen.
ko.lol for desktop
Your champion pool, your matchups and your patch notes, on your desktop while you play.
Free. For Windows.