Weight lifting should be done to failure.

What about intervals?

asking because I just failed miserably on my second 20 min Z4 interval and feel like it was a waste… but in weight lifting it’s what should be done.

Yes, one's last interval should be one where one was not able to hold the power or couldn't get as far up the hill. In this sense, failure is just reaching the drop-off point. That said, I don't think that's always necessary, just when one has, in the past, done enough of said interval so that the next step is to go to failure. Or say one is doing a hard steady-state ride. It's not necessary to go all the way to serious leg pain. Maybe after one has done enough of these things to be in really good shape, go for the pain. No sense in going for the pain when it's just going to put you in the hole for 3 days. Gotta be in good shape first. I think it's the same in the gym. Get in shape first, then go 'til failure. First rule is don't screw yourself up.

Your experience was fine. Of course you couldn't hold the power the whole time on the second one, but good job for trying. Keep trying, maybe once a week. It'll slowly get better and getting better is the whole idea. Can't get "better" without a benchmark which shows what worse was. That said, probably a better idea is to next try two 15' intervals, then go for the 20' when you can do 15'. Or maybe 3 X 10'? It's never a waste - every hard thing one does is a good thing. In strength training, there'll be like 10 reps, adding weight until one can't do 11, not adding weight until one can't do 2.

Why do you think that? What's needed is enough gradual adaptation to reach a goal. If you go to failure, then take time to recover, you often need much more recovery time. The extra recovery time becomes detrimental. Besides that, you wind up getting very sloppy in form toward failure. There's a greater chance of injury. Perhaps to the failure of perfect form and control might be closer to what's necessary.

Look at it in terms of specific reps done with specific weights. If you wind up with more reps over time because of less time off for recovery or injury, chances are you get greater gains.

As far as whether or not you failed, only you get to decide. I say you are doing very well in just the attempt. At least you didn't kill yourself. Sometimes I think it's good to listen to the voice that warns us to back down.

No it doesn't have to be done to failure. Leaving 2-3 reps in reserve is a smart thing to do when training for muscle or strength.... If you're using weights to train for speed and power you definitely need to stay far away from failure and keep fatigue to a minimum. Training to failure often just increases recovery time and the risk of injuries and burnout.

I don't even do strength training to failure...but anyway, context matters in your failed zone 4 interval. Assuming you're fresh enough to be attempting zone 4 intervals, are you riding to a newly established FTP? Have you built your interval times up to 20 minutes?

I don't do FTP tests anymore, but rather just train off of the estimated FTP from Intervals.ICU. When it gives me an FTP increase I'll do my zone 4 intervals at that new power, but at a shorter duration. I'll start out with 3x10's, gradually increasing the interval duration until I'm doing 2x20's or 30's. Once I'm comfortable with those interval durations in shorter rides or indoor sessions, I'll train them deep into a longer 4+hour ride. All that being said, sometimes I just have bad days and struggle to do intervals at my normal power.

To clarify my OP - I'm not planning* to train to failure in the future, nor do I think its the correct thing to do on a bike. It's not a pleasant experience.

Weightlifting was powerlifting for me - we trained to failure to force adaptations. And we adapted.

*My refined question - can positive adaptations happen when training on the bike to failure from time to time? Shocking the system of sorts... or are those intervals/efforts just wasted? I'm getting a little bit of both answer off the interwebs...

I don't do FTP tests. I did do a one hour all out ride for no apparent reason a few months back - full gas. I would think that is better data than a 20 min test. And my training since that point has not really changed or eased up. I don't "train" or particularly ride that hard - 70-75% Z2 or lower, some Z3, some Z4 intervals.

I rode yesterday fresh - figured I do 2x20 at 90-95% of that prior 1 hour power. First interval was easy except for the last min. I figured that was mental. 15 min break at low Z2. Second interval the power was off and I was limited to Z3 - 2+/- min and everything shut down. Complete failure.

It might be beneficial, or you might end up injured. It mainly depends on what you consider failure. I don't doubt that you adapted when weightlifting to failure. I've done it myself, but the new and better way of training is to get in more training over time with less recovery and possibility of injury.

Here is a pretty good summery of what I am talking about.

Context - - - - as it applies to me. When I am strength training in the gym for general conditioning, it is multiple reps and sets somewhere around 80% 1 max rep. If I fail to complete the last couple of reps in the last set, failure will depend on the fatigue I have going into the workout and number of sets.

Strength training to support sprinting is a different calculus.

Sprinting on the bike and training for sprinting on the bike is about max efforts after a prescribed warmup with long recovery periods between intervals.

With respect to 2x20 threshold efforts with 15 minutes RBI, performance on the second effort would be about my initial fatigue going into the workout, consistence of diet and carb fueling during the warmup and first effort and then carb intake for the second effort.

I just returned by Mallorca Spain last night where I did a mountainous route that featured 5 climbs over 25 miles with 2700 feet of climbing. The start of the course is a lead in to a 6.5 % grade. One descends and then starts another longer easy grade followed by a steep descent and then a climb to the top. I had to climb back up the hill I descended at the start of the ride on the return. My goal was to perform as well on the last climb as I did on the first.

I did the harder climbs at threshold, medium grades at SST and false flats and other stuff at VT1/ high zone 2. And of course, sometimes I would just coast along or ride recovery.

I had a solid breakfast start and I packed a lot of carbs to eat on the ride. I forced myself to eat the carbs during the ride and I stayed ahead of the game. The last climb was no problem, I performed well and beat a couple of Germans in some fun cat 6 racing up the final climb and I felt great at the end and did a practice time trial the next day.

None of the riding I did during the mountain ride or TT ride the next day was to failure per se nor did my power fall apart at the end. But I know myself and my capability and I eat a lot.

Finally, with respect to some intervals, yes, they should be done at maximum to get benefit, such as a set of 15 second max power intervals. And yes, the max power intervals will take longer from which to recover. And yes, max power sprint intervals are beneficial and improve not only sprint power but aerobic capacity and yes they require a lot of carbs. YMMV.

My understanding has been that leaving one interval (or a couple weightlifting reps) "in the tank" is a better way to train, because while it provides just a little less stimulus, it greatly improves recovery time, allowing you train again sooner. In other words, the cost-benefit ratio goes downhill in those last efforts right before and during failure.

This is, generally speaking, not correct.
The general rule is that you want to leave one in the tank. I.e. you should be pretty confident you could do at least one more interval. Of course, this is somewhat subjective. If you're doing really long intervals, this may not be the case.
Also, sometimes, but not often, you do want to go absolutely all out.

Easy, your 2x20 efforts with 15 minutes RBI are not at threshold. The typical rest for 2x20 is three to five minutes, and one should not be that gassed at the end. Your fitness today may not be what it was a few months back. To use it as a measuring stick, it only counts if you can reproduce it today. Do a proper FTP test first and see where you stand. I like Kolie Moore's test, https://www.trainingpeaks.com/blog/t...ing-protocols/. Sounds like you're new to structural training, but either way, it might be good to start with 10-minute intervals, with a total session time of up to your time to exhaustion, and progress until you can handle 2x20. For me, it's more of getting into the groove of things, especially after a bump in FTP or after a break. Also, keep in mind that any interval less than 10 minutes is a waste of time.

Saw this in my YouTube feed and thought of this thread.

Steve Magness is a runner, but he has a lot of really informative videos and the principals are all pretty much the same.
Also, wtf? This is totally not true.

Train to failure on every ride you do. No. Train to failure on some rides or for some segments of your ride. Sure. Why not.

Listen from 29:45 to about 34:00

I think the above video pretty much covers how one should do interval training for optimum anaerobic development, except for sprinters of course.

However, there's another form of interval training, not many people are aware of, called Aerobic Intervals. This video describes them, by the same guy in the above video.

Let me refine my first statement about weightlifting to failure.

I didn't mean every session. Lifting to failure at key times or even a 80%-20% type situation (except one tends to go hard in the gym more than 20% of the time) has immense benefits in building muscle strength and power/forcing adaptation. To make the big gains one should go heavy and to failure as a normal part of the overall training plan.

I fully understand the 80/20 +/- rule in cycling - I do 10/15 min intervals, 3-6 min intervals, long sessions of Z3 - but spend the majority of my time in Z2 or less. Probably too much time in Z2.

I guess the root of my question - can going to complete failure every so often, with a solid recovery plan in place, trigger some adaptations that one wont get without pushing that hard?

What exactly are you calling complete failure? Accelerating or climbing as fast as one can until you can't stay at that anaerobic level anymore? And have to slow down to or below your lactate threshold or whatever level lets you continue to ride.

Or are we talking about a 90 minute or so ride ridden entirely as hard as you can and arrive at the finish line not able to do more than pedal a resting or slower pace?

Completely blowing up during an interval for example. To the point where you can't even turn the pedals anymore.

Can you really do that? Essentially you are trying to Bonk. Only when I was a noob to riding for fitness and exceeding my FTP or Lactate Threshold for too long a time have I even come close to the thing that many refer to as a Bonk.

However all one has to do is slow down or shift to a lower gear ratio and continue to pedal. Only on a long steep hill where one doesn't have a sufficiently low enough gear ratio for their power output can I imagine one not being able to continue riding at some lower level of effort.

Bicycles are just too efficient for converting human power into linear motion.


When training to failure with weights... aren't you just lifting the same weight till you can't lift it any more? You could just take off some weight and continue lifting. Similar to shifting gears on a bike.

I suppose on a bike, with a PM, you could ride a a particular power until you can't maintain that power. And I do that quite a bit on the hills I try to best my time on. But when I can't maintain that power, all I have to do is slack off on my effort.

1. apparently you can. You've passed your limits and everything just shuts down. Too soon for a sugar bonk - you just hit it hard. And yes, you can turn the pedals a bit - but in my case it was from 260-270w down to 30-40w. It all exploded in a matter of seconds. You see it in racing when a rider goes full gas to launch the guy and blows up.- you reach your limits, but they still make it up the mountain eventually. VO2 max intervals for example - if you were not able to reach "failure" state, then you could just keep on doing the intervals over and over - but you can't.

2. Weight lifting/training to failure. Typically you are doing say 5 sets with progressively lower weights and lower reps. The last set 2-3 reps at near max and you go until complete failure - failure at that amount of weight. That is how failure is defined for that purpose. Certainly you could pick up a set of 5# dumbbells and do some reps after - but that's not the point. "Failure" was reached.

And in the case of weight lifting, reaching this level of "failure" every so often is where huge gains come into play. You overload the system and it responds. All I'm asking is if the same can happen while training on the bike. Is this level of failure really a negative thing.

Is it a negative thing in weight lifting if you do it every session you lift?

In cycling, I'd think it'd be a good thing to do certain segments till you blow up. But not for every segment and not for the majority of every ride.

Your example of #1 isn't unlike my saying that you just shift to a lower gear ratio and/or just go slower with less power. So I guess we sorta disagree on what failure is. If you make it up the hill or to the finish line of your ride, that's not a failure. Just maybe not a winner. Losers didn't fail, they just didn't cross the line first.

Based on what the two coaches in the YT videos linked here say, the answer is essentially 'no'.

If you're failing on things like curls, or lateral raises, that's no big deal. But, you really shouldn't be regularly failing on things like squats or deadlifts. I'm not aware of any powerlifting program that recommends going to failure on any large compound movement.