An excessively sensationalist title, but I need to keep with this week’s theme.
One of the less obvious things a Skill Build Analysis can do is give us a more complete idea of how a patch change has affected a hero. In 6.80, Legion Commander received these changes:
- Moment of Courage now only procs one extra lifesteal attack
- Moment of Courage buff grace period increased from 0.5 to 1.0 seconds
- Moment of Courage cooldown reduced from 1.2 to 0.9
There’s a lot going on there, but without a lot of complex modeling it’s impossible to know what the net result of the changes is. Luckily for us, we don’t have to model anything. All we have to do is look at her stat pre- and post-patch, and we can have a good idea how the changes have impacted her. And as we’ve seen, DotaBuff’s early results suggests that the net effect is a pretty big nerf and that she was one of the hardest hit heroes in the patch overall.
But Dotabuff results are inexact. We can’t rule out the possibility that some of LC’s decline could be attributable to an environmental shift, and we have no idea how the increased losses are distributed. Perhaps some skill levels or playstyles are more affected by the change than others. Using a new SBA we can look at the win/loss changes divided both by bracket and by skill build, giving us a more detailed picture of the implications of the patch. In the case of Legion Commander, I made the argument a couple weeks ago she’s more successful as a laner than in the jungle, and now with the nerfs to Moment of Courage in 6.80, that argument has received quite a bit of supporting evidence.
The build usage on Legion Commander is mostly unchanged between the patches, so I won’t be including that today. What I want to focus your attention on are the success rates of her builds, primarily the Overwhelming Odds (Q) and Moment of Courage (E) builds.
Across the board, Moment of Courage builds take a big plunge in 6.80, while Overwhelming Odds builds weather the storm far better (aside from High 6.79, which is very likely just sample error). It’s also noteworthy that the overall decline appears to be much more pronounced in the higher brackets, though I hesitate to speculate why this is.
Another way to look at the same data is on a point-by-point basis, and that adds another dimension to the shifts going on. While it’s true that overall build rates haven’t changed dramatically, there is one noticeable shift in how people are building Legion Commander in the 6.80 environment. It was completely unheard of before the patch ( < 1% of players), but in 6.80 ~10% of Legion Commander players in the High and Very High skill brackets are skipping Moment of Courage entirely during the early game (levels one through eight), and these players are actually doing quite well for themselves. Here’s what the win rates looked like prior to the patch:
What you see here are pronounced valley shapes for both Q and E. This basically is another way of showing that Q->W and E->W were the two dominant ways to build Legion Commander in 6.79. Q->W almost always took Moment of Courage as a one-point wonder, whereas E->W did best skipping Overwhelming Odds entirely.
6.80 changes things completely. Q is still a bit of a valley, simply because Overwhelming Odds is a pretty dumb skill to partially invest in, but E is now a straight point-by-point decline, and given the popularity of 0/4/4/1, it’s likely bringing Press the Attack(W) along for the ride.
Maybe Moment of Courage got hit harder than intended with this fix for how the Dota2 engine handles the skill, and we’ll see it brought back up in future patches. On the other hand, maybe this is the new normal that Legion Commander will be balanced around. After all, it’s not unheard of for a Dota hero to have a passive that should never be maxed first.
In any case, I would strongly recommend not maxing Moment of Courage during the early game in 6.80. In the comments of my previous Legion Commander post, someone brought up that she could jungle sufficiently well without maxing MoC, and while I do not have enough experience jungling her to vouch for the efficiency on this, it is absolutely worth looking into if plan on level 1 jungling with her.
More to the point, I would say that these results suggest that there now a very large performance gap between laning and jungling for Legion Commander. Jungling with her might still be worth it if it’s necessary to do so for the continued sanity of your matchmaking acquaintances, but at that point you might be better off with any of the actually competitive jungle heroes. 4/4/0/1 builds putting up a 53% VH win rate in a patch where her overall VH win rate is 44.6% does not paint a positive picture for the viability of her jungle.
And of course you’re still welcome to try to develop new ways to make her jungle work. These stats aren’t proof that it cannot work; merely that most of the people currently trying are struggling. Just be aware that the new patch has dramatically weakened Moment of Courage, and that you’re going to have to find ways to compensate for that if you wish to experience a similar degree of success to your pre-patch expectations.