checked back in to see how the debate continued. Wish I hadn't. Some folks views are so entrenched it's ridiculous. My point remains you can make stats prove anything - I argue they prove everyone starts to fast, OG & PMJ are saying 'cos the proper runners slow down it's alright if we do. Garbage.
Of course, the fast guys don't plan to slow down - they are racing. They are not trying to have a good day and simply set a pb (not that getting a pb is simple!), they want to smash their pb / maximise their position. This means they must take a risk - they hope to maintain pace, but balance of probability (as 'proved' by the stats) is they won't.
OG - I would bet my mortgage on fact that your pb after a pb to halfway would've been quicker if you'd been slower at halfway. But of course, we'll never know - and you certainly won't if you never try to start a bit slower.
Bottom line, we should all run our own race, but I maintain, until you've run an even or negative split it's wrong to say they're not the way to give you the best result. For the longstanding sub315-threaders, I will leave with reprise of my first and last rule of maximising your chances of a successful & enjoyable marathon: don't be greedy.
Evening all, just popped in to see how you're all doing and (a) reminded how many of you there are these days, posting some fantastic training and racing times (b) my favourite subject is being discussed.
Slokey is spot on. Point is until you've done a negative split you can't know if you were too conservative or not. I have and maintain that it's the best way to run a marathon, really enjoyable being the fastest person on the road just by staying the same pace! So I might have left a minute or two "on the road" but I might not. We'll never know.
My advice, which anyone can ignore, is set out at a pace you feel you should be able to maintain, not at one you might be able to.
The bottom line is the lack of even and negative splits at VLM is a result of too many people "planning" to slow down. It's a self fulfilling prophecy.