Item Slam Patch 7.41f
Guardian Greaves

Guardian Greaves: best heroes by WPA

Guardian Greaves in patch 7.41f: WPA of the purchase on 27 heroes, by role and build slot, against the other items bought in the same slot. Highest WPA, still inconclusive: Rubick core, boots upgrade (+0.79 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
RubickCoreBoots upgrade44 (35.2%)+0.79−1.38 to +2.80Inconclusive
Crystal MaidenSupportBoots upgrade42 (6.0%)+0.68−0.07 to +1.41Inconclusive
DisruptorSupportBoots upgrade225 (71.9%)+0.52−0.61 to +1.81Inconclusive
SnapfireSupportBoots upgrade723 (86.9%)+0.29−0.41 to +0.95Inconclusive
Dark WillowSupportBoots upgrade102 (27.7%)+0.20−0.54 to +0.85Inconclusive
LichSupportBoots upgrade39 (10.6%)+0.14−0.90 to +1.01Inconclusive
SnapfireCoreBoots upgrade42 (10.2%)+0.11−1.20 to +1.16Inconclusive
Bounty HunterSupportBoots upgrade177 (5.1%)+0.08+0.03 to +0.14Inconclusive
TidehunterCoreBoots upgrade54 (27.4%)+0.07−0.16 to +0.27Inconclusive
NecrophosCoreBoots upgrade87 (5.0%)−0.03−0.14 to +0.10Inconclusive
MiranaSupportBoots upgrade1,758 (89.6%)−0.11−0.18 to −0.02Inconclusive
GrimstrokeSupportBoots upgrade152 (60.1%)−0.16−2.93 to +2.22Inconclusive
WindrangerSupportBoots upgrade338 (78.4%)−0.20−0.35 to −0.04Inconclusive
Queen of PainSupportBoots upgrade44 (91.7%)−0.33−0.60 to −0.08Inconclusive
Ogre MagiSupportBoots upgrade627 (83.8%)−0.49−0.95 to +0.12Inconclusive
Ogre MagiCoreBoots upgrade33 (37.5%)−0.53−0.97 to +0.03Inconclusive
MagnusCoreBoots upgrade37 (25.2%)−0.58−1.64 to +0.53Inconclusive
MagnusSupportBoots upgrade44 (66.7%)−0.74−2.53 to +0.99Inconclusive
HoodwinkSupportBoots upgrade606 (84.2%)−0.76−1.51 to −0.10Inconclusive
Witch DoctorSupportBoots upgrade591 (87.9%)−0.86−1.33 to −0.38Inconclusive
TechiesSupportBoots upgrade534 (57.1%)−0.94−1.60 to −0.15Inconclusive
RubickSupportBoots upgrade610 (62.2%)−0.99−2.29 to +0.34Inconclusive
PudgeSupportBoots upgrade118 (7.4%)−1.09−2.02 to −0.09Inconclusive
Winter WyvernSupportBoots upgrade314 (66.8%)−1.13−1.94 to −0.24✕ Worse
Shadow ShamanSupportBoots upgrade127 (42.6%)−2.23−5.64 to +1.62Inconclusive
EarthshakerSupportBoots upgrade66 (46.5%)−2.24−5.74 to +1.37Inconclusive
Nyx AssassinSupportBoots upgrade54 (43.9%)−2.29−4.82 to −0.18Inconclusive
Skywrath MageSupportBoots upgrade77 (51.3%)−2.63−4.72 to +0.21Inconclusive
ZeusCoreBoots upgrade37 (25.2%)−5.18−11.54 to +1.08Inconclusive
ZeusSupportBoots upgrade65 (35.3%)−5.59−10.31 to −1.63Inconclusive
UndyingSupportBoots upgrade2,655 (96.8%)––Consensus · 96.8% of purchases
UndyingCoreBoots upgrade155 (78.3%)––No alternative
DawnbreakerSupportBoots upgrade154 (96.2%)––Consensus · 96.2% of purchases
Queen of PainCoreBoots upgrade45 (28.7%)––No alternative
DawnbreakerCoreBoots upgrade33 (28.7%)––No alternative

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.