Comment.

Hi Tim, would you mind sharing your thoughts on pace training for the Ironman marathon. The final run is by no means fast paced, but more about survival/who slows the least. The top pros run around 2.48 (4 min/km) on tired legs, but that pace is far too slow, to improve fitness, for their training pace, except for the long race specific workouts. Targeting the HR would probably present the same problem. I´m a firm believer in good biomechanics, which means avoiding running on tired legs, too much. What was your approach, back in the good ol' days of training three disciplines? Claus Bech, Denmark

These are some quick thoughts that may help..they're a little bit disjointed. There's really nothing secret or unique about what the pros do. I think we were all doing the same thing. The biggest difference between the 2:45 guys and the 2:55 guys, I think, may be how rested they are.

I could answer your question with one very simple and very effective sentence. Simply get as fit as possible on the bike and the run. It really does come down to this 'Zen' like statement. You can't do better than your best. But that's not the answer you want..
There are several key components to running well at Ironman.
First- you need to pick a realistic goal (usually close to an older race best) but an optimistic one. Say you have run 4:00 in an Ironman, I would tell you try for 3:50 (8:46 pace). Taking that goal you then want to work towards the goal of running a fresh MAF test at approximately 30"-45" faster than goal pace for the Ironman. On race day you of course won't be fresh but you will be tapered and will have aid stations. Not implying that you can 'control' your pace by sheer will in a MAF test, but frequent testing will tell you if your goal is realistic. On race day you need to use a mix of pace (time) and HR.. or even just pace. I would tell anyone to train for that optimistic goal then on race day give it a go and click of your goal pace as long as possible.
Next, get fit on the bike. If you put the world record holder for the open marathon in an Ironman, he wouldn't run well.. probably wouldn't finish. This is because he isn't fit on the bike. When I see a guy like Macca run a 2:40 at Kona that says to me that he is an incredible cyclist (not to mention a phenomenal swimmer)! So in order to exploit good run fitness you have got to come off the bike relatively fresh.
The brick. HUGE component to running well at Ironman and also in picking a realistic goal on race day. Bricks are race simulation days, but don't fall in to the trap of performing bricks that are too long. My longest brick I ever did was 6:00 total as: 100 miles on the bike and a 9 mile run. Dave Scott may tell a more experienced athlete to include several bricks of ~80 miles followed by ~13 miles of running with some work done at goal IM pace and even 1/2 IM pace. I can only say that it depends on the athlete and how big their over-all volume is each week. If you're training ~30 hours a week then I would recommend erring on the lower side. A lower volume athlete may benefit from a bigger day. If you can't recover from it.. don't do it. If you're going to miss 3-4 days of training afterwards.. don't do it. One key point to bricks that I believe in is that you need the run to be very race specific. I don't see too much benefit to running off the bike at more than 30" per mile slower than goal pace unless it is a very short (~20:00) transition run to simply get used to running of wasted legs. If you can't hold goal pace then there's a problem and you may be just trying to 'pad' your training log. Save the longer, slower runs for your long run.
Which brings me to the long run. The conundrum in this lies in the fact that you are running fresh rather than after 2.4 miles of swimming and 112 biking. This is where the HR monitor is really a key training tool. Do your long runs at MAF (or ~20 beats below LT) and simply get as aerobically fit as possible. On race day- it will come together if you race your best, execute great race nutrition protocol, and keep your mind from wilting. If you aren't able to run at least half of your long run at goal IM pace then either you are too tired or your goal pace is too fast. For an experienced and fit athlete you should be able to run the last half at goal pace. My long runs were always 3:00. I felt, and still do, that your long run should approach your goal race time. A 3:50 goal means you should be running long runs near 3:50 total time! You often hear elite athletes talk about not running over 2:00 (Frank Shorter) or Ironman guys like Mark Allen say not to run over 2:30. Well, put that in to context in regards to their race times. Frank Shorter ran 2:08, Mark ran 2:40. In a percentage sense- they were running near their respective race times. There is, without doubt, risk in running over 3:30. Risk of injury from breakdown of muscle and loss of support from muscles, ligaments and tendons from fatigue.. but this is Ironman not checkers. Risk is good. Be smart though. There's a fine line between the hero and the idiot.
Rest. Beating yourself down with huge training is good to a point, but if you don't absorb the miles then they will hurt you rather than help you. Setting scheduled rest breaks every 2 weeks or 3 weeks is critical.
If you simply get as fit as possible on the bike. Then get as fit as possible on the run. Nail 3-4 solid brick sessions in the last 8 weeks followed by disciplined rest. Taper effectively. On race day it will come together. The last 8 miles of an Ironman are run 50% with your legs and 50% with your brain. The urge to slow down is overwhelming but you can make the decision to slow down or not. It really is possible to continue holding a pace with simple thought.

A final note: the above is the basics. I would also include a tempo emphasis from 6-12 weeks out from Ironman and tons of hill work through out the entire 20 weeks prior to race day. The Ironman run is about strength so holding a pace that is roughly 12-20 beats below LT during long sections of your long runs or doing a specific medium long run at near your 1/2 Ironman race pace is great for building the muscle endurance required to hold it together late in the race.
A long brick may work for a lower mileage athlete (Claus- you don't want to run on tired legs so this may be good for you)- try doing a 80-100k ride with fractions at 1/2 IM race effort then brick to a 20-30k run with nearly all of it at goal pace.