It’s a bit late, but I do actually read the comments sections and noticed various people requesting a more complete list of the results used in 6.80 Hero Analysis. So here you go.
Sortable spreadsheet including sample size data
Chart of VH-H-N Win rates
This entry was posted on Wednesday, March 5th, 2014 at 1:22 am and is filed under Dota2, Heroes, Links. You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
This is great. Interesting results that disagree with low-end pub wisdom.
Has Farm Dependency changed in 6.80? I want to make a 6.80 hero picker layout.
Thanks! Really interesting to see what heroes are “easy” or “difficult”.
I have a question, I remember seeing a reference (on Reddit maybe?) to some site that surveyed player MMR (obtained via a bot) and matched them to the Normal / High / Very High buckets.
So they got an estimate that Very High was 3800+, and High was 3200-3800, and Normal was anything below 3200. That’s only from memory, now I don’t know what the site was… do you know of any site that’s tried to do a comparison like that?
This would be extremely helpful because most players can’t know their own bracket nowadays (Normal / High / Very High) but they can look at their MMR rating.
Bah, somehow my train of thought got derailed, I wanted to ask if you by any chance remembered seeing that site or reference because it sounded extremely useful (I saw it while I was out browsing Reddit on my mobile phone I think and didn’t think to bookmark it).
It would certainly give a lot more context to these numbers, instead of saying games at Very High look like “this” you would say games above 3800 rating look like “this”.
I haven’t seen that site, but a reliable conversion between bracket and MMR would be helpful given how many people who will now never see the old brackets in action.
I’ve thrown together the usual infographic using just VH and N data for pick rate and win rates. I’m just going to estimate VH as >4000 and N as <3200, from gut feeling it feels right (seems congruent to what I saw that site did, and also jives with everyone on my Steam friends list who was previously N / H / VH and now translated to ranked rating)
One possibly not so minor issue is that this data does not present an accurate representation of pick rates. Previously I took a random 500 matches per day over the course of two months to get a 30k game sample. Recently I changed methods to getting 500 random matches for each hero per day and then purging out duplicates. The upside to this is I can now get 7.5k-11k matches per day and hit a +30k sample size in under a week. The downside is that the usage rates of the sample are no longer a direct estimation to the usage rates of the overall population.
I’ve tried to implement a workaround for this, but my last attempt was bugged in bizarre and frustrating ways. Without that, what I have aren’t true pick rates, though they likely correspond in some way to the actual pick rates.
It’s also of note that this sample was taken during the release of Terrorblade and Phoenix, so their usage rates are exaggerated relative to what they are now.
Thanks for the explanations. The Pick Rates will be skewed, definitely, but I don’t think it’s that bad. If anything the skew should be greatest at the low end where you have artificially inflated the apparent pick rates of the least picked heroes but if anything compared to the 6.77 sample and others done last year the lowest picked heroes are still hovering at around 3% frequency per game, so if it’s being skewed it’s not noticable. That being said I never intended to showcase least picked heroes, since they all seemingly flatline near the same low pick rate.
Yes I’m aware that Terrorblade and Phoenix are going to have above average pick rates.
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.
Feel free to contact me at firstname.lastname@example.org
Enter your email address to follow this blog and receive notifications of new posts by email.
The Contempt Theme.
Blog at WordPress.com.
Get every new post delivered to your Inbox.
Join 175 other followers