I wouldn't bother with another test.... A few years ago I was in "record" mode and I kept a couple of years (say 15 000 miles) worth of consumption figures in a spreadsheet. You'll be lucky if you get consecutive figures within 10 to 15% of each other. Over, say, 10 data points you can start to see trends. Even if you think you're replicating the route, wind, traffic, ambient temperature and even atmospheric pressure will vary your engine's performance....and consumption.
I think some of the other comments are less informed than they might be. On a properly set up engine a vacuum advance - under good conditions - can provide up to 30% improvement in fuel consumption. I live in the country and face almost no traffic lights or other such hassles and my consumption dropped from 30 ish to 40 ish mpg after we fitted the vac advance..... which came from a 70/80s Fraud Escort.
Someone mentioned that they would only work when driving "sedately". True, but consider that an S type engine is only providing about 25% of its capability at 50-60 mph...that's a pretty light throttle. My last trip down the Hume to Melbourne worked out at a bit under 40mpg while cruising at 4K revs (which is a bit on the naughty side of State laws

.
These days getting the A/F ratio correct is a lot easier than it used to be. I have two sets of needles that give almost identical HP/performance... but one uses a fair bit less fuel...
Cheers, Ian