I have solar but no batteries, so a different starting point, but I have done many a back-of-envelope ROI calculation to evaluate whether it's beneficial to switch TOU plans, add batteries or more solar, or shift/shed/add loads. Since I'm inherently lazy, I've never bothered to do the full detailed spreadsheet with Green Button data - but I've worked out several rules-of-thumb that I think can help give you a rough idea.
The basic arbitrage, as others have mentioned, is essentially load-shifting - moving consumption from a peak price to an off-peak price. Except with Powerwall, you don't change your time of behavior, instead you fill the Powerwall during off-peak, and then self-consume from the Powerwall during peak. Thus for each kwh you do this, you will arbitrage (save) the spread - e.g. if peak is 50 cents, off-peak is 20 cents, you arbitrage (50-20 = 30 cents), vs not doing it.
Steps:
0. Calculate the TOU spread.
Arrange the TOU plans in order of increasing spread - E-1, E-TOU-C, E-ELEC, EV-2A. (E-1 is basically a TOU plan with no spread.) Also to note, as the spread increases, the Off-Peak gets cheaper, in addition to the Peak getting more expensive, that's the incentive of TOU. f a plan has tiers, use the higher Tier 2 (since you have no solar, that's going to be almost assured; if you have solar, use baseline i.e. Tier 1). Fortunately, all the plans have the same Peak Period 4-9 pm, and for most, the only relevant spread is in the 4 summer months June-Sept - so use the summer rates and spread. (In other words, the winter spread is <10% of the retail rate, and thus you'd lose money trying to arbitrage due to Powerwall inefficiency).
In summary, the summer peak spread for those plans are presently $0, $0.08, $0.22, $0.31.
1. Baseline your current annual cost on each plan.
In lieu of downloading your detailed Green Button data, go to the Electric Rate Plan Comparison instead. PG&E wants everyone to move to TOU, so they've already analyzed your past 12 months of usage - furthermore, if you click into the details, they will compare the bills on a month-by-month basis. Generally, the higher-spread plans only become economical as you increase your electric consumption significantly (EV's, or large summer A/C usage). If you do see that a TOU plan is significantly cheaper than what you have now, put down your pencil, and switch NOW - what are you waiting for?
More likely, without an EV, you're in the middle where the TOU plans may be somewhat more expensive than E-1. These are your baselines for each plan, already taking into account your off-peak, part-peak, peak consumption. The difference is what you have to make up vs E-1, via the Powerwall arbitrage, to make a TOU plan worthwhile. Also note that E-ELEC uniquely has a $15/month fixed charge, so there's another $180 you have to make up vs the other plans as well.
2. Calculate how much peak usage you can arbitrage.
Ideally with one Powerwall, you will want to shift 10kwh/day x 120 summer days = 1200 kwh. From E-1 to EV-2A - that could give 1200 x $0.31 = $372 arbitrage. Subtract this from the annual EV-2A estimate from the Electric Rate Plan comparison, if it's now cheaper than E-1, it's worthwhile to change plans. You can do this for all the plans to find out which is best. But the question is, do you have 10kwh of peak usage each and every day - if not, that's where you have to go to the more complex analysis folks are suggesting.
But if you do, then it's as simple as above. Some ways to eyeball summer peak usage:
a. Those with solar (NEM, TOU) have an annual true-up that breaks out Peak, Part-Peak, Off-Peak net consumption by month. If it's well above 300 kwh/month, then you likely have 10 kwh/day.
b. If you go into the Electric Rate Plan Comparison details, it compares each plan again your current plan on a monthly bill basis. If the month summer bills are well higher on TOU than E-1, then more of your usage is during peak.
c. Think about your summer usage - do you run A/C, charge an EV, have a welding hobby - between 4-9 pm every day? Then you may have 10kwh/day.
3. Determine whether you're in the one corner case that warrants more detailed analysis.
If you don't have a clear-cut answer already, there is only one scenario that might warrant a more detailed analysis - EV-2A is the only plan that has other large spreads not accounted for above, that are much greater than 10%. Namely:
a. Summer months, part-peak spread - there is a decent part-peak spread from 3-4pm and 9p-12a.
b. Winter arbitrage - there is a decent spread the other 240 days of the year, in both peak and part-peak (same TOU periods as summer).
EV-2A is the only plan that benefits from doing a detailed hour-by-hour spreadsheet analysis, if Steps 0-2 make is close but not clearly better.