"You may re-roll all of the dice on a single test that did not score a hit.

"You" - Subject of the sentence
"may reroll" - Verb of the sentence
"all" - Object of the verb

The original sentence is "You may re-roll all".

Both "of the dice" and "on a single test" are propositional phrases that complement the noun before it.

"You may re-roll all (of the dice (on a single test))."

All what? of the dice
Which dice? the one's on a single test

However, "that did not score a hit" is a quantifier for a noun that is quantifying "all". Even if it was modifying a preposition, it would modify "dice" not "test"

"You may re-roll all that did not score a hit."

If I were going by the "that" clause quantifying "test" then you would not be able to spend edge to re-roll dice on a test unless you achieved 0 hits since it would be "single test that did not score a hit."

However, I believe you and I are interpreting that line identically.

If I roll 10 dice and get 3 hits and 7 non-hits. I can spend a point of edge to re-roll just the 7 dice that did not obtain a hit. (<-- My Interpretation) You were leading, perhaps inadvertently, towards a situation that a test with a single hit was ineligible for edge to be spent to re-roll dice.

EDIT: Okay. I am really bored apparently.

You and I both come to the same conclusion, I was just pointing out where the opposition was coming from.

You and he and most people. The only other way to interpret the sentence is that any of the dice rolling a hit on a test disqualifies the roll from being able to be rerolled in any form, which is fairly absurd.

Well, having taken multiple Linguistics and English courses in college, I'm pretty aware of the fact that an ambiguous sentence such as the one in question in this thread can be diagrammed in multiple ways. Indeed, that sentence is pretty badly and vaguely written.

I lean towards the interpretation where you can't reroll failures if you get a single hit for two reasons:

a.) It matches up with the rules for Skillwire Expert Systems and what they state.
b.) I've seen how overpowered it is with large dice pools, both as a GM and a player. If you have lots of dice already, then it almost ensures you get a ridiculous number of hits on whichever test you choose to use the Edge on. There's almost no incentive to use the Edge dice in the pool or after the roll, since the chance of rerolling hits on missed dice is mathematically superior to hoping your 6s explode.

Just my thoughts.

I wasn't paying attention earlier, but is it *really* better to reroll non-hits than to roll your whole (DP + Edge) with Ro6? It just seems wrong. For one thing, you're not getting + Edge, and then Ro6 on the whole thing… *shrug*

If you throw 21 dice and get 7 hits, you can either have thrown edge before (+6 dice) or throw non-hit dice (+14 dice). Which is better:

14 / 3 = ~5
6 / 3 = ~2 +~1 die reroll

?

Statistically, adding edge first means your dice pool is "1 hit in 2.5 dice" rather than "1 hit in 3" whereas rerolling failures gives you 1/3rd of 2/3rds (2/9ths) of your dice as more hits.

What if you threw 27 dice and got all sixes? (IMG:style_emoticons/default/biggrin.gif) I'll need better data than *that*.

Statistically, adding edge first means your dice pool is "1 hit in 2.5 dice" (with a slightly increased number of dice) rather than "1 hit in 3" whereas rerolling failures gives you 1/3rd of 2/3rds (2/9ths) of your dice as more hits (for a total of ~5/9ths average hits, ratio of 1:1.8 ).

I see. And how does it affect things if Edge is very large or small, relative to the DP? I wouldn't call changing a DP from 8 to 12 'a slightly increased number of dice', for example.

If Edge is large, compared to the dice pool, the extra dice you get will trade in (at 1:2.5) for more hits than you'd get by trading 1:1.8.

Example:

10 dice, 6 Edge:

16 dice -1:2.5-> 6.4 hits, average (2.4 from Edge alone)
10 dice -1:1.8-> 5.55 hits, average (2.22 from reroll)

10 dice, 3 Edge:

13 dice -1:2.5-> 5.2 hits (1.2 from Edge)
10 dice -1:1.8-> 5.55 hits (2.22 from reroll)

Thanks! I'd have to make some graphs, but it seems like there's a fine line between the two options, depending on DP and Edge.

There most certainly is!

Including if the number of hits you need is greater than your base dice pool. (IMG:style_emoticons/default/wink.gif)

Even if rerolling failures is statistically better, if all hits still results in a failure (or death, when dealing with damage) then it's better to add Edge beforehand!

The basic math has already been done.

I said, I wasn't paying attention earlier. (IMG:style_emoticons/default/wink.gif) What I really meant was that that list is very unhelpful. Thanks again, Draco.

While it is true that the sentence as written can be 'correctly' interpreted either way linguistically (you may reroll a test that did not score a hit - or - you may reroll all dice that did not score a hit), it cannot be done so within the Shadowrun game, as I demonstrated earlier, although possibly poorly.

A test is qualified as either a Success or a Failure.
A die is qualified as either a Hit or a non-Hit.

A test cannot score a 'Hit', and so cannot be the target of the modifier "did not score a hit". As such, the modifier must be applied to dice.


"On a single test, you may reroll all dice that did not score a hit" is the valid interpretation, Rules as Written.
Anything else is a House Rule.

The internet; Where three people can argue the same side of an argument while claiming that everyone else is wrong.

@Muspellsheimr:
I see where you are coming from, certainly. But I'd be giving way too much credit to the writers of the rules if I thought they were that nuanced and careful in their uses of the terminology. Truth is only the person who authored the statement knows what was really written.

Edit: If they'd written the statement as you did ("On a single test, you may reroll all dice that did not score a hit") there would be no ambiguity or room for interpretation.

Did they really pay that much attention to get the rules to function as intended? Hell no. We need only look at Indirect Combat vs. Object Resistance (among many others) to see that.

This is probably one of only a few of the convoluted instances that Rules as Written actually matches the (probable) Rules as Intended. Nonetheless, it is in fact RAW.

Still, quite possibly the best use of Edge, as far dice efficiency, remains using it to gain an extra IP.

A good point there. And it goes great with re-rolling failed dice. (IMG:style_emoticons/default/biggrin.gif)

I know of at least one author, possibly more, with sufficient understanding of grammar to merit that credit. That's not to say more don't, just I've only had one stress the importance of the proper use of the semi-colon.

That said, I read the statement and interpreted it to mean refill dice that were not hits, without registering the possibility of another interpretation. So I ask you, did you initial come to the same conclusion prior to grammatically analyzing the statement or was analysis require to understand the statement?

Grammatical analysis is entirely suspect, anyway. (IMG:style_emoticons/default/smile.gif)

Tests can score hits if they want to. I'm much more convinced by the argument that it would be bad to limit the use of the reroll to zero-hit situations.

D = Dice in DP
E = Edge

Expected # of hits from an exploding 6s die = 2/5

Expected hits from adding Edge after the roll = (D/3)+(2E/5) = (5D+6E)/15
Expected hits from re-rolling failed dice = (D/3)+((2/3)x(2D/5)) = 9D/15

Equality occurs if: 5D+6E = 9D or 3E = 2D

So, if the DP is less than 1.5 times the Edge rating then adding the Edge dice is preferrable, but re-rolling failures is preferred otherwise. This will often be the case, but then shouldn't it be true that Edge does you more favour when applied to something you're experienced at? (Bear in mind that Edge is more akin to resourcefulness than actual luck).

The bolded and underlined part is in error. Should be (D/3)+((2/3)x(D/3)) as you don't get the Rule of 6 on the rerolled dice.

Erm... yes you do. Check your copy of SR4.5 again. Page 62, I believe.