asgelle

There is a big difference in the inertial load from flat riding to climbing. That is, the neuromuscular demands for how those W/kg are transferred to the bike are different. That isn't to say the carry over isn't there; the great majority of improvement in W/kg will transfer, but that doesn't mean there isn't some degree of specificity for the two cases.

chasm54

OK, some experience on actual hills can't hurt and will probably help. Some people have never spent time out of the saddle, for example. But the OP asked if he could improve his climbing by training on the flats. He certainly can, and if he does it intelligently he'll get most of the gains without ever seeing a hill.

Heathpack 

I ride a lot in the dark. A hill that's tough in the daylight always seems easier in the dark when I can't see the top. Or maybe I am just distracted by how cool it is to ride with bats.

I don't remember the original question and I'm too disinterested to look back. IMO the most efficient place to improve your power is on the trainer, because you can hit every interval target exactly every time, terrain never dictates that you be over or under or take more or a rest than is ideal.

However your best training is always by doing the thing which you are training for, don't you think? For example, when climbing a hill at X watts, you will be going slower than when you are producing that same power on the flats. So you won't be cooling efficiently, it will feel different and may necessitate that you modulate your power output differently. If you are on a very steep hill, you may be required to climb at or above threshold and you won't be able to give yourself a mini-break short of stopping (and then it will be tough to re-start on the hill). Lots of little differences like that.

If you can't train on hills all the time, sure you can work on stuff off the hills that will make you climb better. But whenever possible, I say get theeself to an analogous hill and practice the actual thing that you're trying to get better at. If its a long sustained climb, then my approach would be to get myself onto a long sustained hill whenever possible.

chasm54

Hey, listen, if the hills are there, ride them, of course. But in the OPs case they aren't. Can he improve his ability as a climber without hills to climb? Yes.

Sure, it's different. But sticking the bike in a big gear and churning along in Z3/Z4 at a cadence of around 70 makes it a reasonable approximation. Just need the determination not to back off, because the hill wouldn't allow you to do that.

EDIT

Turn it around, HP. Do you think all those miles in the hills have helped your fitness to ride a flat 20k TT? Those activities are different, yes. But the fitness translates from one to the other...

Heathpack 

Well OP lives in the Bay Area. He has access to hills. The real question is whether it's worth the effort to get to the hills to train. To me, the answer would be yes, at least semi-regularly. I can't say for sure that's what OP should do, because I have insufficient insight into his life to know for sure how big a hassle it would be. But it would be more ideal.

Do hills make me better at flat TTs? Sure they do. My biggest problem is finding flat terrain to train on, so I will ride an 8 mile TT type effort up a hill. Negative splits, steady power, threshold effort. That's my practical compromise. But I still put the bike in the car a drive to flat terrain to practice TT efforts. Why? Because it's different enough that it's worth it to me. Producing steady, even power on a steady, even 5% grade is pretty straightforward, the hill kind of forces it. But even the tiny undulations on flat terrain make steady power output more challenging, at least for me. I'm also typically at a higher cadence on flat terrain, so the way in which I breathe is different and the effort seems more aerobic to me subjectively, even though the power is the same. It takes way more concentration on the flats and it's easy to lose your focus when your brain is a little addled and the terrain is not forcing X watts out of you. It's a different task in the fine details.

So yes, I can work FTP and improve power and endurance either on hills or flats. But if I'm trying to improve my riding on the flats, I make the effort to get myself to flats. I'm kind of meticulous in how I prep for things, I want to get every little detail down pat. Of course not everyone is like that. In which case flat efforts can be used to sub for hills and vice versa. It's doable for sure, even if not 100% perfect.

Drew Eckhardt 

Research suggests otherwise.

Riding above your lactate threshold improves it more than riding at or below it. 10 minute intervals make you faster on 1:15 hill climbs than 75 minute efforts or 30 minute intervals which are closer to the event.

Riding below your aerobic threshold in Coggan Z2 increases the pace you can sustain above it better than actually riding the Coggan Z3 pace fast riders use for centuries and Ironman races.

Heathpack 

Not sure where you got the concept that practicing climbing by actually doing climbing excludes the possibility of doing intervals. Do both. Or at least that's how I do it.

Also, I don't think it's accurate to say that "research" has proven that the 10 minute long interval is the one best way for every cyclist to improve his or her hill climbs. It depends greatly on what that persons natural abilities are, where he/she is in a training program, what his/her current weaknesses and strengths are. There is no research study that could ever be designed that would prove what is the best approach for every cyclist period.

I ride 30 second intervals, I ride 1 min intervals, I ride 3 min intervals, I ride 5 min intervals, I ride 10 min intervals, I ride 20 min intervals, I ride 30 min mock TT efforts, I ride 40 min tests, I ride long 1 hour climbs (sometimes multiple climbs in a single ride). I even intentionally ride in zone 3 for large portions of entire workouts. Gasp.

sprince

Yes, generally the research quoted isn't asking the right questions when it comes to training on the flats vs. climbs. Obviously improving your power output will improve climbing to some degree, but that's like saying you can improve your running by running faster. And it should also be noted these would be marginal gains at best. The research subjects are always identical subjects, or at least as identical as the researchers could find. There is little or no variation in build, weight or natural ability. The research basically says that identical subjects who have already spend some time climbing tend to have nearly identical results regardless of the type of training done in the course of the study in preparation for the test climb. More interesting would be a study group of people that haven't done a long climb in the past year. Have half the group train for 4 weeks on climbs and the other half train on the flats and then put them all on a 3,000 ft climb. I think I know how that would turn out.

My personal experience is that nothing replaces the real thing. Even a few long climbs make a drastic difference in my uphill speed if I have not been on a real climb for some length of time.

Heathpack 

I guess I would not say that a research project is not asking the right questions so much as a properly designed study is going to be testing a very specific hypothesis. The question being asked would not be "what is the best way for amateur cyclists to improve their climbing?" It will do something like test the hypothesis that in a given group of specific test subjects that A type workouts will result in better power output (or faster time up a hill or whatever their endpoint is) than an identical group of test subjects doing B type workouts. I can guarantee you that the group of test subjects is first of all not going to be huge and second of all not be terribly similar to me personally- they're going to be young, male, possibly avid cyclists but not necessarily.

The research project may conclude A type workout did in fact result in improved power output after their 6 week (say) test period. But that does not mean someone who is different from their test subjects or who has longer than 6 weeks to train or who is not exclusively doing type A workouts would have the same results.

If you want to interpret the literature you have to be extensively familiar with it. You have to know enough about the subject to understand the flaws in a project, whether the results really support the conclusions of the paper. It helps to know the individuals in the field, to know their biases. That's just the reality of evaluating scientific literature. It's why people in the biz sit down regularly in an endeavor called Journal Club. You get together with your colleagues and really read & critique a few journal articles and discuss what they paper really says. Some of them are well done and every conclusion is adequately supported. Some of them are terrible and their conclusions are disregarded.

I actually read in a general way about training and physiology but I recognize that's not my area of expertise so I don't take it too far. Who figures out my workouts and my training schedule? My coach. Because that's what he does for a living and he has a much broader understandng of the subject than I have.

I definitely think improved power basically means better climbing. But beyond just looking at power, there's small things you can learn by climbing a lot- about pacing & generating power on a hill. Nuance stuff, not huge differences. I'm into nuance, so I say climb on hills when possible. For me, doing a few climbs will not make me significantly faster- just a little better and more on top of my game. Increasing power (which as I said I personally think is best done on a trainer) or decreasing weight are 90% of the battle. It's up to each of us whether the final 10% of the battle is worth seeking out by training on actual hills.

gregf83 

Anecdotally, many people claim to be able hit better power numbers on a hill than on the flats. The inertial load while climbing is lower than riding at higher speed on the flats. Inertial load is also lower on a trainer which again anecdotally, mainly claim to hit lower numbers than riding outdoors. Personally, I find it easier to generate power with a tailwind which has high inertial load vs a headwind.

The studies I've looked at don't seem to show a significant difference in power with inertial load changes. Perhaps the difference is only psychological.

Seattle Forrest

Most people don't have a power meter, they're expensive.

Hills are a pretty reliable way to get cyclists to produce a lot of power. It's hard to pace yourself, and while you can put out as much power on the flat as you do on a climb (like when you sprint), most people don't, at least for much time. Hill repeats are good work/rest interval workouts, even with no measuring devices at all they're going to force you into or near the right zones. But with a power meter, you can do the same work going down hill if you want to.

There's a place in Seattle (HSP, on Nickerson) that will rent you a power meter for $75 for a week. When I started using one, it was eye-opening.