Winner Stats for Stone Wardens

[grooog]

There isn't a ton of info out there for Stone Warden players. And even the best-written, most helpful guide is subjective. So I collected stats from ten players who've recently beaten the game using a Stone Warden. (I'm fixing to get my first win by playing my stone warden as carefully as possible and by emulating what these players chose.)

see the spreadsheet here: http://goo.gl/zmqgrY

The first tab is pretty much just the raw data for side-by-side comparisons, and the final column shows you the average for the ten players. (For example, players typically had around 60 base magic, so the average in the rightmost column says 59.3)

Other tabs at the bottom of the spreadsheet are different ways to look at the options chosen by these 10 sample characters. How often did players put at least one point into Pulverizing Augur? How often did they put 5/5 points into it? Which infusions were the most popular? Which escort rewards were chosen most frequently?

Prodigies are pretty straightforward, and I see no need to do any fancy math with them: almost everyone chose Spectral Shield and Eternal Guard. A couple people took Irresistible Sun. One person took Crafty Hands and one person took Draconic Body.

I was not surprised that stone wardens love taking Unflinching Resolve, but I was surprised that no one took the whole Conditioning line. I guess SWardens have better uses for their category points.

Anything surprise you about the results? Do you have thoughts about different ways we should look at these numbers? Are you a spreadsheet wizard who'd like to access the sheet and add a tab for your own number-crunching? Sound off!

[Sirrocco]

I note that the skill numbers for short stuff make little sense, especially as compared to, say, greta. Short Stuff looks like he's maxxed just about everything. Did he hit the Infinite Dungeon after winning or something?

[grooog]

I note that the skill numbers for short stuff make little sense, especially as compared to, say, greta. Short Stuff looks like he's maxxed just about everything. Did he hit the Infinite Dungeon after winning or something?

I was already thinking about halving the weight of the two players who had zero lives left, since they nearly didn't win. Upon further review short stuff has a crazy number of talent points and the "adjustable exp" addon, so I'll just remove him, possibly to later replace him with someone else or just leave it at 9 players for now.

[breadsmith]

Thanks for this, as it's probably the closest thing to a guide for what is clearly the best class in the game.

[jotwebe]

As a general caution, be careful about using averages. If you're looking for a typical example of a population, they can be very misleading if there are two (or more) very distinct variants, for example a melee build and a ranged build. It's generally a better idea to use the median for something like this, but even that's problematical here. Pretty interesting comparison though.

[grooog]

Yeah, the average is a really unhelpful metric. Maybe I'll just take that out. I find it more interesting how many people took at least 1 point in a skill, and how many people maxed it out.

When you control for race AND class, there's less wild variance, but there's definitely a difference between people who go Deeprock and those who don't.

Here's what I did find interesting:
Most people either invested heavily into Stone Vines, or not at all. But it was a pretty even split, so I feel it's safe to conclude that Stone Vines are a matter of taste, neither ballast nor must-have.

Similarly, players who took Stone Spikes or Deeprock Form tended to invest heavily when doing so. No one took both, so either one seems to suffice as a ranged option. The Stone tree was the most unpopular, but then again, Batu, the only player with zero deaths, went 1/1/1/5 for Crystalline Focus.

Earthen Power and Eldritch Shield were highly popular across the board. So was x/5/0/0 Stone Alchemy, as usual, but I didn't include Generics because 1) more work, 2) most of the race/class generics are no-brainers, and 3) non-race non-class generics (escorts and such) will vary depending on what you get in your game.

[jotwebe]

Yeah, the average is a really unhelpful metric. Maybe I'll just take that out. I find it more interesting how many people took at least 1 point in a skill, and how many people maxed it out.

It's great when you want to know how many people you can put on an airplane or something like that.

When you control for race AND class, there's less wild variance, but there's definitely a difference between people who go Deeprock and those who don't.

Here's what I did find interesting:
Most people either invested heavily into Stone Vines, or not at all. But it was a pretty even split, so I feel it's safe to conclude that Stone Vines are a matter of taste, neither ballast nor must-have.

Similarly, players who took Stone Spikes or Deeprock Form tended to invest heavily when doing so. No one took both, so either one seems to suffice as a ranged option. The Stone tree was the most unpopular, but then again, Batu, the only player with zero deaths, went 1/1/1/5 for Crystalline Focus.

Earthen Power and Eldritch Shield were highly popular across the board. So was x/5/0/0 Stone Alchemy, as usual, but I didn't include Generics because 1) more work, 2) most of the race/class generics are no-brainers, and 3) non-race non-class generics (escorts and such) will vary depending on what you get in your game.

Yeah, looks like Batu was a maximal physical damage melee build.

Interestingly my most successful build so far isn't represented (http://te4.org/characters/4384/tome/2f6 ... cf3cf45007), basically a Stone Archmage with Vines.

[grooog]

Hey, that's really cool. Congratulations on a roguelike win! *Edit: er, sorry, I assumed instead of reading, but cool that you had the balls to try roguelike on a meleeish class though!

If you want to add other game-winners to the spreadsheet, feel free to use the button in the top right to request access to the sheet, I only stopped at ten to keep my workload low. Ideally I'd like the "weight" of each character to be inversely proportional to their number of total deaths, but my spreadsheet-fu is a bit novice for something like that.

*I added the character You Asshole, You who did manage a roguelike win, though it was way back in game version 1.01.

[SageAcrin]

As a note of how misleading figures can be sometimes; This is my own Stone Warden, which had no deaths, and did moderate investments in two of the categories(Spikes and Vines) mentioned.

Admittedly, Vines wasn't very good at the time, but it did get somewhat buffed since then. (The main differences for old vs new is that Spikes has gotten buffed a bit in 1.05, and that Vines/Stone got buffed a while ago.) All told, I'd still probably go for moderate investments on both.

But in general, if you look at any sheet, you will notice a habit of 1-or-5 on skills. Probably a holdover from Diablo 2/3 mindset(one of the most popular games with a levelable skill system, and one that rewarded the middle levels very poorly.), as moderate leveling is often the best idea.

Stone is more of a victim of playstyle tendencies than actual power, I think. People tend to not think in terms of spells running without staffs, and the huge statistical split(A class that has uses for all stats to some notable degree is unusual) will probably mean players avoid Magic as the easiest stat to ditch, early on.

Still, statistics are good. Just tossing some random comments in.

[seattlesid]

Wow. Impressive. THANK YOU!

[Tasidar]

I only play roguelike and an earth mage stone warden is my only win.