I have a question about whether I can expect to run Boston faster than I did in my qualifying race (Oct 17).  I started training and was more rigorous about following a plan this time.

I have been using a Pfitzinger 18/55 plan fairly consistently.  I have had some trouble with too many slow miles in weeks of higher volume.  For example, the last Sunday long run with a specific tempo is 18 miles with 14 miles at marathon pace--not only did I fail the 14 miles at marathon pace but the total time for the 18 miles was discouraging.  Note that in January I ran a half marathon at a bit faster than marathon pace and could have easily gotten to 14 miles at marathon pace.  I am doing all the prescribed interval and stride work in the Pfitz plan.  This weekend I changed the Pfitz plan slightly (the plan was 10K race plus 3 mile cooldown on Saturday and 17 long distance miles on Sunday) to run a 5K with just 2 miles of cooldown and run 20 on Sunday).  My run on Sunday was not spectacular but I felt steady the whole time (except the 300 ft climb at 18.5 mile point).

I was feeling a little more hopeful about getting a desired Boston result but then I looked at my long runs prior to my qualifying race and realized I have been running consistently slower on my long runs (slower on most of the shorter runs too)  That is, my best long run was 3 weeks before the race and it was 22 miles at 25 seconds slower than race pace.  My 20 mile run yesterday was 50 seconds slower than race pace (and my longer runs up to this point have been generally worse).

Last time I ran only to get enough mileage during the week and focused on doing some tempo running on my long runs.  It was fairly flexible, I ran more often but shorter distances during the week.  It was based a lot on feel (and life constraints) but looking at the mile times it was clear that I would have some nice tempo runs a few times a week when I was "feeling it".  I took more days off for minor issues so only got over 50 miles twice and over 40 miles for the last 6-7 weeks.  Pfitzinger has "tired" me out more (even with 5 days of running) as I am running a few 10-12 mile runs during the week.  Also, the last few weeks the plan has over 10 on Friday, 6-9 on Saturday before the 17+ run on Sunday.  I certainly understand why my long runs are slower--I guess the issue is that I have never completely trusted in a plan before so will the speed issues work themselves out in the taper period.

So that is why I am curious if anyone has enough experience or intuition to guess at whether I am headed for a (slightly) better time or whether I should scale back a bit on expectations.  It just seems that if I am running nearly a minute my goal in training it will be hard to turn that around in the final race.

I can't imagine running a 22 mile long training run at 25 seconds slower than your race pace for a marathon.  Maybe if you are running 6:00 miles I guess.  Even then that seems awful close to race effort to me.

My last marathon cycle I ran an 18.7 mile run at 9:38 pace 6 weeks before the marathon and a 18.6 mile run at 9:08 pace 4 weeks before the marathon.

I had quite a bit of quality work in there for me and ran a 1/2 marathon race 3 weeks before Boston.

I ran Boston at a 7:57 pace and that was into a pretty stiff headwind last year.

My take is if you are getting in the work you are going to be fatigued and those long runs might be a little slow.  I always measured where I was at by my marathon pace tempo runs mid week on the Hansons plan.

I feel that weekly mileage is a better indicator for a marathon than just about any other metric.

My 2 cents.  Pretty hard to tell without seeing your training log though.

I have done your plan a few times, and I failed that 18 w/14@MP run more times than not.  That and the longest tempo run are both very hard.  If you don't have perfect weather, it's easy to have trouble on them.  I think the best predictor of success is the tuneup races... hopefully you'll do one or two of them.