It is. But the ticks you're seeing are the damage over time. So when you add up the Ignite damage you have to divide it by 2 ( the interval that which Ignite ticks) And you get the damage your combat log shows.

Except that it's (very nearly) as likely that you'll lose the pyroblast ignite to the refresh munching since the scorch/pyro will land relatively close in time to one another, and in any case, even if you ensure that the fireball or the scorch doesn't eat the pyro's ignite (if it crits), you're as likely as not going to lose it to your living bomb or existing pyro dot.

Add to that that scorch if you already have hotstreak up is a pretty meh thing to do with a gcd (since most of the reason a scorch rotation isn't a horrific dps loss comes from increased frequency of hotstreak procs), and its highly unlikely that you'd pick up damage (though certainly there would be times where you would, just less frequent than the dps loss you'd invest to generate them).

Question on glyph choice;
Due to the low levels of crit, (I'm running at 27% self buff) I'm finding that the extreme infrequency of hot streak procs means that even in extended fights I'm not getting many pyroblasts off, however living bomb is ticking away the whole time. Wouldn't a 3% increase to LB be much better than a 5% crit increase to something as rare as Pyroblast! ?

Take into consideration that the glyph, used with MA, increases the critical strike chance of ALL spells by 2%, making it the best prime glyph. Glyph of LB can be considered situationally for fights that you will have LB constantly on more then one target. For single target fights, glyph of LB falls very short of the other ones.

The idea of replacing Fireball with Pyroblast had been looked at in the past, and it was shown to be a DPS loss. I just simmed it again, using both my character's current gear as well as the 372 BiS profile in SimC. In my gear, it shows as a ~300 DPS loss, but in BiS gear it sims as ~1k DPS gain.

There is a caveat though. There are very few fights that allow you to stand there and free-cast, and the longer cast time of Pyroblast spam would most likely negate this gain. There is no reason to deviate from the current accepted casting priority.

I love this resource, however I have some questions/comments on the way stat weights are presented here.

A lot of people on other message boards consider these weights to be absolute Truth and believe that they're valid at all gear and stat balances. Of course, this is very much not true. Linear weights make the simplifying assumption that all the marginal value of increasing any one stat is constant and that the interaction between different stats is negligible. Actually neither of those assumptions are true, so the linear weights only give good approximations near the baseline while the secondary effects are still minor. A note that stating what stats were used as the baseline for the numbers would be useful. I know that it was the T11 372 set, but I can't tell if the base was "naked" or after a best guess of appropriate gems and enchants had already been added.

On a side note, maybe it'd be more appropriate to publish the 359 gear numbers instead, since nobody who has a full set of 372 gear is going to need a website to tell them how to gem/reforge.

It's especially interesting to hear that RAWR is sometimes telling people to reforge crit back to haste in extreme cases. This might be a bug, but it could also be a reflection of diminishing returns for crit.

The OP does have this note as it links to Simulationcraft Results which is BIS 372. I have to disagree that it is inappropriate to provide stat weights for gear which the intended audience will never wear. The simple reason is administrative: we cut down on a lot of warnings/infractions for people who would otherwise come here to post "wut gear should i wear". It's also rather irrelevant that the stat weights are for different gear than they are wearing. An arbitary 359 gearset would also give different weighs to what they are actually wearing.

The issue with Rawr "reforging to haste" is usually when it finds some breakpoint. For example if Rawr sees that an extra 10 points of haste will give an extra tick of combustion under bloodlust, then it will happily reforge one item for 80 crit -> 80 haste, and on another item, in a seeming contradiction 70 haste -> 70 mastery, all the while telling you that crit is 30% better than the others. This because it only needs the 10 haste to beat the 80 crit but otherwise the mastery was better. I expect lots of people are looking at those types reforges as bugs or glitches, but remember that with default settings Rawr is doing exactly as you tell it: 300 seconds fight, no interruptions, no movement, bloodlust at the perfect moment, combustion at perfect moments with volcanic pots during molten fury, etc.

The latest Simulationcraft Results now include correct ignite munching. Combined with the additional mana cost reduction on Fireball on the PTR, this results in a huge shift in stat priorities:

Intellect=3.1468
SpellDamage=2.4011
HitRating=2.2689
HasteRating=1.6104
CritRating=1.2664
MasteryRating=1.1239

Int lost some value (because Scorch weaving will no longer be necessary), and if we can trust Simcraft, Haste is now our best stat. What are your thoughts?

Taking the new manareduction of fireball into consideration, combined with a look at the "Mana Timeline", which leaves us at 20% of our mana pool at 450sec, i think that makes sense. Since Scorch weaving isn't required anymore, we need more haste to get our mana pool to 0% before the fight ends. Furthermore haste isn't directly affected by ignite munching.
Maybe we'll see another change in stat weightings, when we get enough haste to emtpy our mana pool before the fight ends.

I'm always really wary of using stat weights to determine a "best stat". As someone said previously, these statements tend to get taken out of context (BiS 372 gear), and interpreted as "stack X".

The way I see the results, is simply that the changes to ignite modeling indicate that the best idea is to balance your stats. The previous BiS set has always been crit heavy at the expense of haste and mastery. The new results seem to confirm the general conclusion of the Rawr solver, which is that haste and crit are more or less equal in value, and it is optimal to increase both ratings at a roughly equal rate, as haste increases the value of crit and vice versa. The new simcraft numbers can be seen as a reaction to the crit heavy BiS set, and the resulting increased value of haste due to a skewed gear set. Of course its hard to say anything conclusively until theres a new BiS set, and we can see what the ratio of haste:crit:mastery is in that gearset.

Also, the stat delta for stat weights is set to 300 in that simulation, which seems rather high, given the large number of haste breakpoints present during a fight for a fire mage (combustion, living bomb/pyro/FFB, heroism temporary buff). Perhaps something smaller might be better suited to avoiding potential discontinuities in the value of haste.

The change seems twofold to me. First in the more accurate modeling of Ignite, naturally reducing the value of crit, and second in the change to mana costs. So far in cataclysm we've had to account for going out of mana. We've been all over this before: haste will improve the individual dps of any given spell, but also reduces dps if it has a negative effect on a rotation. Up until now haste could force us to spend more time casting a suboptimal spell to prevent us from running out of spells to cast completely before the end of a fight. If the fireball cost change is so great that we are no longer under this constraint at all then I think simcraft is just showing us that we can get more from haste because the gear set in question (and possibly the spec, looking at 3/3 clearcasting and 2/2 improved scorch) is not held back by scorch cycles.

Barring some major error in the model, the thing to find out I guess is potential levels of haste that do require scorching. I'd love to see a revised BiS set according to the new listed stat weights run through simcraft. If no such level exists and haste still rises above then it will be the dominant stat, otherwise there will be a more complex, iterative approach to gearing where we achieve some relative levels of stat balance.