How dumb is the MVP system?

General Feedback
It's this dumb: http://imgur.com/a/hMUKa

Apparently, your death count is the only thing that matters. Great news for anyone who wants to sit out and just chew bubblegum.

To be fair, the healer was awesome at keeping us alive, but they didn't kill three forts and two keeps and they didn't do most of the wave/merc clearing while repeatedly pushing off or destroying the enemy team.
Imo healers shouldn't get MVP than often, let's be real, healers don't have the carry and outplay capabilities. Some peel, some have CC, some can be agressive, but most of the time, it's not game changing and don't deserve MVP because of healing done and 0 death
You don't get to kill forts, get mercs, clear waves without the healer keeping you alive.

It was also 39 takedowns for Auriel (most in the game) vs your 29 takedowns, meaning she was everywhere helping win teamfights.

Healers don't often make flashy plays, but they are the most important role. The biggest rule in QM matchmaking is that you can't have a team with support against a team without, just because it is such a big advantage.
10/09/2016 01:21 PMPosted by Ghazshakul
Apparently, your death count is the only thing that matters. Great news for anyone who wants to sit out and just chew bubblegum.


Nope, but it counts a lot. And thats a good thing.

Usually you can tell who is a good player and who is a bad player by looking at the death counter. A player that is alive contributes so much more to the team in terms of presence alone (soaking EXP, enemys scared to do objective / gank / boss XY) and so on. A dead player is always extremely bad and often enough the reason of falling behind in some way.
Additionally huge numbers like heroe damage can be missleading. Often you see the feeder being #1 in heroe damage ? How can this be ? Thats an easy question. Bad players take fights all the time and therefore fights they shouldnt take aswell. That they end up #1 in heroe damage while the better players only take 2-3 important and winable fights is the result of their missplays.

One single death in lategame can throw an entire game. You are gone for 60 seconds and depending on the map your team is forced into a 4vs5 fight or deff at base which will probably end up in another death by the time you respawn. So throws often start with a single death in lategame that turns into staggered deaths until you lost. Therefore Blizzard wants to benefit players that die less.
I suggest you check the numbers again. The healer had one death, I had two - whereas everyone else on the team had 4, and everyone on the other team was bucking for 10. Both of my deaths happened in midgame, and both were part of a full party wipe (one of which our healer managed to escape).

You really think that's a valid basis for discrimination? Because from where I sat the only difference in our death count at that point was a matter of luck. And considering our relative performance on every other metric, that's basically putting pennies ahead of the dollar.
Stats are misleading though. Siege damage is both minion clearing and structure damage, although the game tracks them individually. That's why there's a reward for structure damage done at the end of the game. It doesn't include minion clearing, otherwise it would automatically go the most siege damage done.

There are a lot of other items evaluated as well. Shortly before I got MVP earlier today as Xul, someone on our team criticized me for low hero damage. In reality I had fired off a couple altars (Towers of Doom) and gotten a couple of camps, had the highest teamwork (most Kill+Takedowns), and was generally owning the map. I only died once and I only had 2 killing blows. But overall I knew I was MVP and that it didn't show in the score screen at all. So sometimes it's hard to tell, but I'm guessing if you watched the replay and focused on Auriel, you'd be impressed by the consistent output on behalf of your team.

On the other hand, the MVP is generated by an analytic metric, so it does make mistakes. Sometimes I question the nominations, but at the end of the day they're meaningless fanfare anyway.

Also, I couldn't find your replay on HotsLogs, so neither team uploaded it yet. Was it from today?
Yesterday, but I'm not in the habit of uploading replays.
10/09/2016 01:21 PMPosted by Ghazshakul
To be fair, the healer was awesome at keeping us alive, but they didn't kill three forts and two keeps and they didn't do most of the wave/merc clearing while repeatedly pushing off or destroying the enemy team.

The main reason Auriel won MVP was because of the total hero kills. Sgt Hammer achieved 29 kills (11 kills, 18 assists) while Auriel achieved 39 kills (7 kills, 32 assists). That difference of 10 kills combined with 1 less death and the healing score compensated out all the extra experience, hero and siege damage that Sgt Hammer did.
The basic formula for MVP goes like this:

Kill + Assist - (Time Spent Dead/Game Length)

Therefore:

Auriel: 7 + 32 - (Time Spent Dead/1102 seconds)

You: 11 + 18 - (Time Spent Dead/1102 seconds)

I don't know the exact time spent dead on both of you unless I watch the replay, but it should be obvious Auriel died once, while you died twice, so you have longer dead timer than her. With the value adding up (Auriel has 39, you have 29), she gets the MVP without question.

Source:
https://www.reddit.com/r/heroesofthestorm/comments/53av50/fyi_how_the_mvp_is_chosen/

10/10/2016 10:35 PMPosted by DrSuperGood
10/09/2016 01:21 PMPosted by Ghazshakul
To be fair, the healer was awesome at keeping us alive, but they didn't kill three forts and two keeps and they didn't do most of the wave/merc clearing while repeatedly pushing off or destroying the enemy team.

The main reason Auriel won MVP was because of the total hero kills. Sgt Hammer achieved 29 kills (11 kills, 18 assists) while Auriel achieved 39 kills (7 kills, 32 assists). That difference of 10 kills combined with 1 less death and the healing score compensated out all the extra experience, hero and siege damage that Sgt Hammer did.


This sums it up
39 kills and 1 death vs 29 kills and 2 deaths. That is pretty much all this thing cares about.
10/11/2016 02:56 AMPosted by Goozmania
39 kills and 1 death vs 29 kills and 2 deaths. That is pretty much all this thing cares about.


You shouldnt spread such false informations, seriously.

I already got MVP as Rexxar with moderate siege damage, moderate heroe damage, moderate tanked damage, moderate kills, moderate exp gained but taking objectives all the time and doing many, many camps.

I am pretty sure the algorithm for MVP is more complicated. Why should it gather data for awards like structures, objectives, merc camps and so on but not take the gathered data into account for the decision of MVP ? The system uses all the data, transforms it into a final overall number (lets call it IMPACT NUMBER) via an exchange rate that was defined for every data gathered. So not the person with the best kill / death ratio gets the MVP. The person with the overall best IMPACT NUMBER - a number that is the combination of everything but not shown anywhere - gets MVP.

Join the Conversation

Return to Forum