Pistoleer Archive

I'll post the data tomorrow, but I finished the test with Ali'qui and ran a SS/FS test with Nezlok. The gist is:

- Pistol Accuracy While Standing seems to stack properly.

- DT/FS/SS/BS1 all seems to have +50 accuracy (same as BS2).

- Kip-up shot has +40 accuracy (definitely lower than the DT line).

No real surprises from finishing set C and set D, except possibly that all the specials have very similar accuracy bonuses. I would have sworn that SS had a better accuracy rating during the testing, too...the numbers don't agree with that, though.

Great thread, Noules. Over time I see this data being as important to us as the speed/damage data. Good work.

---

Noules000 wrote:

- DT/FS/SS/BS1 all seems to have +50 accuracy (same as BS2).

- Kip-up shot has +40 accuracy (definitely lower than the DT line).

---

This is a huge datapoint. I'll start using double tap more often now.

Noules, this is unbelievably good work. Outstanding additions to our understanding of the mechanics behind this game.

One question, has anyone had a miss against a lair, even at range. I can't recall that I have ever missed against a lair? Is this everyone else's impression?

Measured/expected hit rates (hit rates include active defenses):

C5: 44.1%, 18.2% + (double-tap bonus)

C6: 42.6%, 18.2% + (bodyshot 1 bonus)

C7: 25.4%, 23.2% (this is slightly on the high side, but this set was smaller than normal, and within margin of error- just to check for any gross abnormalities between the break)

C8: 39.7%, 18.2% + (kip-up shot bonus)

D1: 22.4%, 25.7%

D2: 47.7%, 25.7% + (fan shot bonus)

D3: 49.0%, 25.7% + (stopping shot bonus)

Running through, we get 25.9% for double-tap and 24.4% for bodyshot 1; there's about a 1.7% margin of error for 200 trials, however, so they're both well-within error of 25%, which is what BS2 seemed to be. Kip-up has a bonus of 21.5%, which is outside of the expected error from 25%. This means kip-up accuracy bonus should be +43 +/- 3, which I arbitrarily pegged at +40.

There's a discrepancy in D1. Not only is the measured accuracy lower than expected, it's lower than Ali'qui's autofire trials in set C. Since both of them have the same base pistol accuracy, are using the same gun at the same range against the same target, my conclusion is that creature to-hit +10 provides more of a bonus than pistol accuracy while standing +15. From this set, it looks like the bonus from Pistol Accuracy While Standing is halved again, so +15 actually provides +7.5 accuracy (which in turn only provides +3.75% chance to hit). We have to use the measured accuracy for the relative comparisons for the special moves, so the base chance to hit for D2 and D3 should be 22.4%.

We then get 25.3% for Fan Shot and 26.6% for Stopping Shot. SS is just within the margin of error from +50, although +55 might actually be a better fit. I don't know the precision used for the style bonus, but for most practical purposes FS and SS seem to have the same accuracy bonus, which we'll call +50, same as DT/BS1/BS2.

So far the only special that hasn't had +50 to accuracy was kip-up and warning shot (+40 and +25).

Other thing to note is 2 active defenses (1 dodge, 1 counter) in 1500 trials. Being prone might hurt the percentages, but I'm suspecting that the -base- to hit percentage controls the active defense rate - lower the to-hit percentage, less chance of active defenses firing. The only data set that doesn't follow this pattern is the 95% capped to-hit set, but I'm guessing the capping has something to do with that. I'm inclined to say active defenses are bugged when the to-hit chance isn't capped.

---

Dinian wrote:

---

Noules000 wrote:

- DT/FS/SS/BS1 all seems to have +50 accuracy (same as BS2).

- Kip-up shot has +40 accuracy (definitely lower than the DT line).

---

This is a huge datapoint. I'll start using double tap more often now.

---

Well, let's do the math...

In terms of raw damage/delay, kip-up is about 11% greater DPS. In order for DT to be a better choice, the accuracy boost has to be at least 11% (relative). Assuming the difference between the two shots really is +10 accuracy, the absolute percent difference between them is 5%. So the cut-off is:

(x+0.05)/x = 1.1111...

or x = .45. This means you have to be missing more than half your shots with kip-up before DT becomes better (neglecting speed).

Fan shot is obviously always better than both.

Note, I didn't say I'll spam double tap... but when I'm running around in circles at 60m, I'll take the accuracy.

Though I'm better off just using bodyshot2 in that case. So much for thinking our specials had any use.

Good work factoring in creature-to-hit. Glad someone has the patience to test all this out.

You wouldn't happen to have a BH handy would you? It would a neat comparison to test them out too. I suspect the to-hit bonus for their specials are lower.

---

Aerius wrote:

Good work factoring in creature-to-hit. Glad someone has the patience to test all this out.

You wouldn't happen to have a BH handy would you? It would a neat comparison to test them out too. I suspect the to-hit bonus for their specials are lower.

---

One of my guildies is a BH/pistol hybrid, but I haven't seen him on in a bit. It's also getting harder to find volunteers to take 2K+ shots a night and the corresponding wounds and BF.

It wouldn't be -too- surprising to find out that pistol specials have the best accuracy, given that accuracy seems to be a major complaint from carbineers and riflemen but rarely from pistoleers. This might reflect the range mods more, though.

Noules,

This is excellent work. If you need volunteerCrash Test Dummies I'd be willing to set one up on your server some time and just prop him up somewhere you can shoot him. Maybe for the running-in-place-in-house test or something where you likely won't find someone excited to run straight into a wall for hours while getting shot.

Let me know.

-Psquire

Noules, wait, in your initial data sets, it seems that your target gets hit MORE when he's grouped with you than when he is not.

So wouldn't that mean SL defense is working backwards?

---

EasyMcRhinopants wrote:

Noules, wait, in your initial data sets, it seems that your target gets hit MORE when he's grouped with you than when he is not.

So wouldn't that mean SL defense is working backwards?

---

Hmm, not sure which you're comparing, but A4 and A8 are the relevant sets (control variables standing, standing, pistol, test variable is grouped/ungrouped). A1 is me using a rifle, and I'm a master rifleman compared to only master marksman with a pistol. I hit a lot more in A8 than in A4.