In this short discussion, I'll math up the difference between the Spellsurge Enchant and the Soulfrost Enchant. Soulfrost gives a bonus of 54 shadow and frost damage, whereas Spellsurge gives each spell you cast a chance to proc an effect that gives 100 mana to everyone in your party over 10 seconds. One of the main points of Spellsurge however, is that the proc rate is listed on the enchant in-game as a 3% chance, however many people on the Elitist Jerks and WoW forums believe that the proc rate is 15%. The one thing that's known, however, is that the internal cooldown for spellsurge is 50 seconds.
I'll be using a typical 10 minute fight as the baseline and for spell cast counts, I'm using the simulator at leulier.com. The main purpose of that simulator is actually to approximate uptime for a warlock's Improved Shadowbolt talent, but since Shadow Priests also do shadow damage (strangely enough), the simulator is designed to build a shadow priest's damage timeline as well. Honestly, it's the best sim I've found for SPriests. It will actually spam MB and SW:D when they're up without breaking a MF that's casting or letting a DoT stay down for longer than it takes to recast.
According to the sim, a Shadow Priest on all-out DPS rotation will, in 10 minutes, cast:
39 Vampiric Touches,
24 Shadow Word: Pains (Shadow Words: Pain?),
72 Mind Blasts,
42 Shadow Word: Deaths, and
112 Mind Flays
For a total of 289 casts.
With Spellsurge at 3%, 289 casts will garner 8.67 (about 9) procs, which, assuming good timing, doesn't hit the internal cooldown at all and can return 900 mana to everyone in your party. At 15%, roughly 43 casts will proc the effect, however with a cooldown of 50 seconds, only 12 procs can possibly occur in 600 seconds, with at most 11 actually being useful since a proc at the very end of the fight isn't all that helpful.
Soulfrost is a bit more complicated, naturally. According to WoWwiki, the coefficients for Shadow Priest spells are as follows:
VT - 100%
SW:P - 146.66%*
MB - 43%
SW:D - 43%
MF - 59%
*The listed coefficient for SW:P is 110%. However, this 110% is split evenly among 6 ticks, so when the Improved SW:P talent is included (adding two more DoT ticks), the total coefficient can be found by taking the old coefficient, dividing it by the normal number of ticks, and multiplying it by the talented number of ticks. 110% / 6 * 8 = 146.66%.
Extra Damage over course of 10 minute battle:
VT - 54 * 100% * 39 casts = 2106 damage
SW:P - 54 * 146.66% * 24 casts = 1901 damage
MB - 54 * 43% * 72 casts = 1672 damage
SW:D - 54 * 43% * 42 casts = 975 damage
MF - 54 * 59% * 112 casts = 3568 damage
For a total of 9247 extra damage from Soulfrost alone. At VT's 5%, that's an extra 462 mana for everyone in your party.
Now, according to the maths, Spellsurge is the better enchant for a good mana battery (hint hint, =P), returning (at the 3% rate) 900 mana, almost twice as much as Soulfrost alone. However, I have a few problems with the Spellsurge numbers. That is, Spellsurge is a chance. There's no gaurantee that it will proc even relatively close to the end of the 50 second internal cooldown. The cooldown forces the enchant to not re-occur for at least 50 seconds. After the 50 second cooldown is up, there is a chance that your spells will proc the effect again. If the 3% chance is true, there is actually a 20% chance that the effect won't even proc again in the next 50 seconds.
Soulfrost, on the other hand, while will not only return 462 extra mana to everyone in the party, but also increase the total amount of damage done by OVER NINE THOUSAND. Which is utterly rad.
However, there is actually a compromise. According to some very reputable sources, you can configure CasterWeaponSwap to recognize and swap weapons for Spellsurge procs. Now this is pretty amazing. What this means is that you can have two weapons, one with Spellsurge and one with Soulfrost and the mod will automatically switch to the Soulfrost weapon as soon as Spellsurge procs and then swap back to the Spellsurge weapon when the CD is up, repeating ad infinitum. I haven't actually tried it because I just found out this was possible, but I'm going to go ahead and download and install this mod now and check it out. According to Shadowpriest.com and Elitist Jerk forum-goers, it's worth it if you have another weapon lying around, and I have a Mindblade in my bank that I just upgraded, so... =P
Thanks for checking out my first Maths post. Depending, I might actually post more. =P