Item Slam Patch 7.41f
Monkey King Bar

Monkey King Bar: best heroes by WPA

Monkey King Bar in patch 7.41f: WPA of the purchase on 19 heroes, by role and build slot, against the other items bought in the same slot. Best confirmed result: Juggernaut core, 2nd item (+4.17 pp).

By hero and slot

WPA of the item against the other items bought in the same slot, in matches of every rank. Slots with fewer than 30 purchases are left out.

Hero Role Slot Purchases WPA 95% CI Verdict
JuggernautCore2nd item171 (1.2%)+4.17+1.40 to +8.00✓ Better
Night StalkerCore4th+ item98 (0.8%)+1.84−3.02 to +7.25Inconclusive
SlarkCore3rd item115 (0.8%)+1.68−2.03 to +5.38Inconclusive
SniperCore2nd item130 (2.8%)+1.65−2.05 to +4.90Inconclusive
SlarkCore4th+ item860 (3.1%)+1.43−1.33 to +3.69Inconclusive
WindrangerSupport3rd item52 (1.0%)+1.21−2.50 to +5.02Inconclusive
JuggernautCore3rd item789 (6.3%)+1.08−1.39 to +3.28Inconclusive
SniperCore3rd item274 (7.1%)+0.85−1.98 to +3.65Inconclusive
WindrangerSupport4th+ item59 (1.9%)+0.65−3.42 to +4.79Inconclusive
LionCore4th+ item80 (2.7%)+0.63−4.46 to +6.83Inconclusive
SniperCore1st item49 (1.0%)+0.27−4.67 to +4.69Inconclusive
JuggernautCore4th+ item1,867 (7.1%)−0.24−1.99 to +1.61Inconclusive
Outworld DevourerCore4th+ item61 (0.2%)−0.53−5.46 to +3.68Inconclusive
Wraith KingCore3rd item39 (0.9%)−0.73−4.63 to +3.18Inconclusive
Drow RangerCore2nd item87 (0.3%)−0.84−4.57 to +2.93Inconclusive
SniperCore4th+ item513 (9.0%)−1.54−4.00 to +0.74Inconclusive
Drow RangerCore3rd item160 (0.6%)−1.80−6.24 to +1.87Inconclusive
Drow RangerCore4th+ item337 (0.7%)−2.10−5.27 to +1.14Inconclusive
Dragon KnightCore4th+ item213 (0.6%)−2.12−7.14 to +2.37Inconclusive
Wraith KingCore4th+ item297 (3.5%)−2.22−5.81 to +1.50Inconclusive
LinaCore3rd item95 (1.1%)−2.49−7.00 to +0.94Inconclusive
Legion CommanderCore4th+ item302 (3.2%)−2.56−6.32 to +1.03Inconclusive
WindrangerCore2nd item150 (1.5%)−2.95−5.45 to +0.03Inconclusive
LifestealerCore4th+ item1,811 (8.7%)−3.07−4.96 to −1.41Inconclusive
Legion CommanderCore3rd item46 (0.6%)−3.36−9.12 to +0.97Inconclusive
Anti-MageCore3rd item57 (0.5%)−3.50−6.71 to −0.33Inconclusive
LifestealerCore3rd item264 (2.2%)−3.76−7.25 to +0.02Inconclusive
MagnusCore4th+ item32 (0.5%)−3.76−10.37 to +2.87Inconclusive
Phantom LancerCore4th+ item40 (0.2%)−4.00−9.49 to +1.37Inconclusive
SnapfireCore4th+ item59 (0.7%)−4.24−8.36 to +0.30Inconclusive
Drow RangerCore1st item33 (0.1%)−4.25−9.38 to +0.06Inconclusive
SvenCore3rd item145 (1.0%)−4.28−8.29 to −1.13Inconclusive
WindrangerCore4th+ item418 (4.7%)−4.35−7.02 to −1.53Inconclusive
WindrangerCore1st item32 (0.3%)−4.88−9.19 to −1.09✕ Worse
LinaCore4th+ item381 (3.1%)−4.97−8.26 to −1.46Inconclusive
Anti-MageCore4th+ item1,066 (5.8%)−5.02−7.12 to −3.19✕ Worse
WindrangerCore3rd item348 (4.7%)−5.42−8.64 to −2.74Inconclusive
SvenCore4th+ item3,284 (8.1%)−6.49−7.62 to −5.22✕ Worse
LunaCore4th+ item596 (1.3%)−6.49−9.35 to −3.46✕ Worse

How to read

What is WPA?

Win probability added: how much choosing an item changed the chance of winning, in percentage points, compared with what the hero usually buys in that same slot and situation. +1.0 means the choice won 1 percentage point more than the typical purchase. Unlike raw win rate, it accounts for the game state when the item was bought, so an item bought from ahead doesn't look good just because the game was already won.

Build slots

1st, 2nd, 3rd and 4th+ count completed items. Components that become another item in most matches (Perseverance, Dragon Lance, Kaya) do not take a slot: the item they build into does. Boots get a card of their own.

Confirmed or not

The ✓ Better or ✕ Worse badge only appears when the effect passes the false discovery control (10%) in both ways of measuring: at the start of the decision and at the moment of purchase. Without a badge, the sign of the WPA is a trend.

Where the data comes from

Ranked (All Pick) matches parsed by OpenDota in the current patch. One model estimates the chance of winning from the game state; another, the effect of each item at the time of the decision. It is association, not causation: the model sees the game, not who is playing.