Item Slam Patch 7.41f
Phase Boots

Phase Boots: best heroes by WPA

Phase Boots in patch 7.41f: WPA of the purchase on 26 heroes, by role and build slot, against the other items bought in the same slot. Best confirmed result: Pudge support, boots (+2.37 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
LionSupportBoots34 (0.2%)+4.41−4.50 to +12.87Inconclusive
Dragon KnightCoreBoots62 (0.3%)+3.90−4.40 to +14.13Inconclusive
SvenCoreBoots136 (0.9%)+3.43−3.48 to +9.60Inconclusive
Nyx AssassinCoreBoots77 (5.5%)+2.95−3.45 to +10.74Inconclusive
PudgeSupportBoots3,550 (9.3%)+2.37+1.05 to +3.94✓ Better
MagnusCoreBoots412 (5.6%)+1.99−2.54 to +6.52Inconclusive
DawnbreakerSupportBoots138 (25.4%)+1.93−2.64 to +6.97Inconclusive
WindrangerCoreBoots96 (0.8%)+1.52−4.15 to +7.64Inconclusive
TidehunterSupportBoots130 (42.1%)+1.47−2.38 to +5.29Inconclusive
WindrangerSupportBoots140 (2.4%)+1.30−3.77 to +6.91Inconclusive
Bounty HunterCoreBoots119 (60.7%)+1.08−1.51 to +3.61Inconclusive
SlarkSupportBoots52 (6.7%)+0.91−5.13 to +7.76Inconclusive
JuggernautCoreBoots2,506 (16.5%)+0.84−0.80 to +2.36Inconclusive
UndyingSupportBoots404 (4.7%)+0.80−2.75 to +5.25Inconclusive
Bounty HunterSupportBoots301 (1.9%)+0.29−4.35 to +4.92Inconclusive
Legion CommanderCoreBoots9,453 (95.3%)+0.26+0.07 to +0.43✓ Better
LifestealerSupportBoots43 (97.7%)+0.17+0.05 to +0.32✓ Better
Night StalkerCoreBoots8,955 (83.3%)+0.12−0.21 to +0.55Inconclusive
PudgeCoreBoots17,029 (88.1%)+0.11−0.11 to +0.32Inconclusive
UndyingCoreBoots2,419 (84.1%)+0.07−0.50 to +0.66Inconclusive
Ember SpiritCoreBoots13,097 (93.2%)+0.01−0.21 to +0.18Inconclusive
SniperCoreBoots35 (0.7%)−0.00−7.76 to +9.13Inconclusive
DawnbreakerCoreBoots13,399 (98.4%)−0.02−0.10 to +0.05Inconclusive
TidehunterCoreBoots9,533 (97.1%)−0.03−0.13 to +0.09Inconclusive
Spirit BreakerSupportBoots15,720 (92.4%)−0.07−0.25 to +0.10Inconclusive
Spirit BreakerCoreBoots1,456 (95.5%)−0.18−0.35 to +0.01Inconclusive
AxeCoreBoots11,819 (98.1%)−0.23−0.33 to −0.14Inconclusive
Wraith KingCoreBoots4,878 (90.1%)−0.25−0.60 to +0.16Inconclusive
Legion CommanderSupportBoots82 (94.3%)−0.45−0.80 to −0.13Inconclusive
Night StalkerSupportBoots69 (71.1%)−0.55−2.43 to +1.28Inconclusive
Ember SpiritSupportBoots107 (90.7%)−0.57−1.15 to +0.00Inconclusive
Wraith KingSupportBoots46 (75.4%)−1.10−2.82 to +0.62Inconclusive
AxeSupportBoots244 (59.5%)−1.36−4.24 to +1.09Inconclusive
EarthshakerCoreBoots1,173 (14.7%)−1.41−3.89 to +0.79Inconclusive
Ogre MagiSupportBoots32 (0.6%)−1.74−7.77 to +4.34Inconclusive
EarthshakerSupportBoots35 (0.8%)−2.20−6.97 to +3.05Inconclusive
SlarkCoreBoots126 (0.7%)−2.66−9.08 to +3.84Inconclusive
Ogre MagiCoreBoots612 (38.2%)−3.25−5.73 to −0.16Inconclusive
SnapfireCoreBoots80 (1.5%)−3.95−12.14 to +4.67Inconclusive
NecrophosCoreBoots458 (3.6%)−5.70−10.14 to −1.19Inconclusive
Spirit BreakerSupportBoots upgrade78 (12.1%)−7.96−13.66 to −2.26Inconclusive
Drow RangerCoreBoots37 (0.1%)−10.79−18.06 to −2.83✕ Worse
LifestealerCoreBoots14,738 (99.2%)––Consensus · 99.2% of purchases

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.