Duel of Wits Strategy (with research!)

Point Point Point. So it gets brought up a lot and some of you are probably already rolling your eyes. It often strikes new players that in Duel of Wits scripting Point Point Point is an undefeatable optimal strategy. It is roundly denied by veterans with great experience. And yet there is a kernel of truth inside the problem. I decided to do SCIENCE and see for myself. I wrote up some VBA macros in excel to run Duels of Wits to compare various scripts. The result is not appropriate as a tool for playing the game - you can’t spend artha, you don’t see the results of individual rolls, etc. But it does do what I intended, which is to see how Point Point Point fares against Obfuscate Rebuttal Rebuttal (and other actions) depending on the size of the dice pools and who has more dice. I did two kinds of tests: best-of-ten DoWs with various strategies pitted against Point Point Point; and super-DoWs starting both sides off with dispositions of 1000 and comparing how much the winning side had left over at the end.

First lets get out of the way a couple of the reasons not to script Point Point Point BESIDES winning (or getting a good compromise). First, advancement - different actions test different abilities and you only keep one test per ability, so you should vary your actions to get more tests! Second, roleplaying - sometimes you’re mad as hell just want to lay down an insult and it doesn’t really faze you that your incite is doomed to failure. Third, it’s more fun for everyone if you make yourself vulnerable and take risks! The main advantage of Point Point Point is that it’s SAFE - you’ll never get feinted out of your action, you’re difficult to incite, and even if you lose you’ll probably get a good compromise.

Now on to the results - everything here is from the perspective of one who is ignoring the previous paragraph and is only concerned with results in THIS duel RIGHT NOW.

So if your opponent is NOT scripting Point Point Point, should you? No. Because your opponent is scripting things like Obfuscate and Rebuttal, Point is solid but you will get more out of throwing in a few feints or even incites while keeping up an offensive strategy including many points.

However, what if your opponent is scripting Point Point Point. Can you stop him without stooping to his level? The answer is that it depends. If you can muster more dice than him then scripting obfuscates and rebuttals and prolonging the debate will help you preserve your disposition (proved this: with 7 dice against 4, you’ll end up with about 20% more disposition left if you use Obfuscates and Rebuttals over Points). If he can muster more dice than you though, then you’d be fighting a losing battle and would be better off scripting Point Point Point yourself and taking solace in the fact that it is your opponent who is scripting badly and you’ll probably end up with a decent compromise. These two situations in my research accorded well with my expectations. If your skills are weak, Point Point Point is perhaps your best hope against an opponent you know will also be scripting Point Point Point.

What if you’re evenly matched? The interesting answer is that it depends on how many dice you are rolling. When both participants rolled 4 dice each for each test, the Point Point Point player won 8 lost 1 and drew 1 against a Obfuscate Rebuttal Rebuttal player. When both participants were rolling 7 dice per test, Point Point Point player won 4 lost 4 and drew 2, although he had the better compromises. So if your opponent is scripting Point Point Point and you are both rolling lower numbers of dice, you may want to consider matching him. It’s a Nash equilibrium if you know a little game theory. If you are both rolling big fistfuls of dice, you can rest assured that actions like rebuttal and obfuscate are decent alternatives (even at very high dice pools I found that Point Point Point had the advantage, but the advantage was quite small at 6 dice or over).

I did some tests with the other actions too. Mostly they confirmed my suspicions, but here are the results:

When you have a skill advantage, Obfuscate and Rebuttal are roughly equally effective against Point. I did tests with Obfuscate Rebuttal Rebuttal and with Rebuttal X3 and got near-identical results.

Rebuttal: Very strong if you have the dice advantage. If you don’t it’s still good but you have to be sure not to allocate too many points into defense - every success in your defense pool that doesn’t stop damage to your disposition is a success that could have been knocking down your opponents’.

Obfuscate: Again, very strong with the dice advantage. It benefits from the fact that +1Ob is a bigger penalty than +1D is a bonus. It suffers from the fact that there is no benefit to having a large margin of success - every success past the one that penalized your opponent is simply wasted. Very good against rebuttal because you’re rolling all your dice vs his defense pool alone - odds are in your favour.

Feint: Obviously if your opponent rebuts a lot, this is great. Vs Obfuscate it’s just like Point. Rather a one-trick pony, but when the opponent is using rebuttal to capitalize on his bigger dice pools, this is exactly what you need to take him down.

Incite: Incite vs Ob Will is pretty bad. It mostly just fails and you take a bunch of damage. If your opponent is using points and dismissals, this will get you nowhere fast - unless you’ve got the skills to hit Ob 4, you’ll just end up with a big penalty to your next action. Incite as a vs. test however is pretty good! It’s still a gamble and it’s meant to be, but if your opponent is rolling avoid, obfuscate, or feint then you’re rolling your dice against theirs and if they win they get some minor benefit at most while if you win then you can unleash your undefended point or dismissal in the next volley. Rebuttal doesn’t get to test against this either, so if your opponent likes to obfuscate and rebut, you should consider throwing in an incite.

Avoid: Useless against dismiss, blocks feint, blocks rebuttal. Pretty self-explanatory. Strong defensive play if your opponent isn’t trying to dismiss or incite you, and it’s also the only way you’ll be getting Will tests here outside of Beginner’s Luck. Useful at times, but you’ll never win an argument simply by avoiding the topic!

Point I already feel like I discussed pretty well - you’ll never be taken advantage of like feint vs rebuttal or dismiss vs avoid, and it’s the best way to do damage. Use lots of these! A strong defense includes defensive maneuvers like obfuscate and rebuttal and avoid, but if you don’t script points then you’re making yourself a target for incites and feints.

Dismiss: Always fun, always powerful - use it as a knockout punch and if your opponent mounts a strong defense you might need to spend some Artha to avoid getting drool on your fancy clothes. Remember that if you beat Obfuscate then you don’t just do margin of success damage, you do ALL your successes in damage.

1 Like

Interesting and useful. Thanks!

The main problem with a Point-Point-Point strategic, as I see it, it’s that I don’t believe many players can think in too many ways to attack the rival argument. If you can’t think in a new argument, in a new and interesting way to beat the rival point of view, you can’t test. And if you repeat yourself, then you begin to suffer dice penalties.

If you want to defend your argument, your position, then you must script Rebuttal. And the players who feel attacked will want to defend themselves very soon. Son, in practice, the player who want to do that strategic (Point-Point-Point) will encounter too many problems in finding new ways to say the same thing over and over. At least in my experience.

Interesting post, by the way.

Thanks, Jim!
This matches with my experience. When I’m teaching new players the DoW rules, I inform them that they must make Points to win.

This quote is the best part, in my opinion: “If your skills are weak, Point Point Point is perhaps your best hope against an opponent you know will also be scripting Point Point Point.” Because here we get into imperfect information and meaningful decisions. If have good skills, I have an incentive to plow you under with P/P/P, but you know that. And if I know your skills are low, I know your best defense against my P/P/P is also P/P/P. This information changes my strategic options. Now if I can lead you to using P/P/P my best options are O/R/R. But if you cotton to that, then you can risk a P/F/P exchange. Knowing that…I need to think more along the terms of O/P/R or R/O/P.

Anyway, I love how the information informs the decision-making.
Thanks for posting!
-Luke

I do love that figuring out what your opponent is going to do gives you a big advantage.

I’d like to see how I could try that kind of experiment myself with a Mouse Guard conflict.

I ran Dam Beavers the other day and the players have since asked me if we can continue with the game making it into a full camapign. :slight_smile:

I was really pleased when the players restored 4 disposition with a well played Defend (vs my attack) and improved the compromise. I was pleased because I was worried that they might think that defend is worthless and that’d devalue feint until the only logical strategy is Attack Attack Attack. It’d be much like one of my BW players who has assimilated Point Point Point as his go-to strategy.

Be interesting to have a little automated tournament, having each strategy compete against all the others. (i.e. in a ‘total points of BoA surrendered over 20 matches against each opponent’ sort of thing). The success metric might be BoA won minus BoA lost.

Because of the rock-paper-scissors element, the tournament design would influence the winner - at the very least through the distribution of opponents (some scripts are much less likely than others; it’s better to be prepared for popular scripts than for rare ones). You could mix in ‘fake’ opponents with higher or lower skills as well, so that’s a factor.

The highest-scoring half would advance to the next round, and so on. That might be enough to weed out the silly scripts early, and ensure that the competitors are being graded mostly by their success against other successful scripts.