Today is the Day You've Been Waiting for (We don't have Alberto's reply) but the LRC Better Running Shoe Site Is Here

Can you add a filter for medications taken?

Bless you, boys!

Very helpful, thanks.

We have posted this link on the homepage now and next week will have a lot more on the shoe site after this crazy weekend of track dies down.

We wanted to get this out now for our shoe geeks.

Thanks for everyone who has submitted feedback.

I didn't see many minimalist shoes. Did I do something wrong with the filter?

Is this supposed to work? Selecting the filters in your shoe finder (e.g. picking Nike) does not actually filter the reviews. Also, there is no way to undo a selection.

LOL at Nike having the 2 best shoes. Nike running shoes suck. I guess they pay the most.

Agreed! Nike shoes aren't the best. But they work. And considering there's a lot of track runners on this site (a small assumption) Nike may be the most worn/marketed. Being from the NW, I think Brooks shoes are much better, but I'm surprised Asics isn't on the top of the list. Asics...are also the most expensive shoes...

so free = better ?

I like where this site (shoe site) is going! I think it will help promote up and coming brands....that seem to keep popping up. Altra's for example are great for those with wide forefoot's but I'd imagine it's hard for a college athlete with a budget to go around experimenting on shoes. Way to go Let'srun!

There's too much emphasis in it right now on number of reviewers. These Nike shoes aren't even the highest rated, but their large number of reviews is giving them a higher ranking. That's silly.

I like this a lot. Thanks.

What happens if all of my recommended shoes are discontinued? Should I quit? In seriousness, you should be able to filter are shoes that are discontinued.

And on the switching analysis, you shouldn't be recommended shoes that are discontinued. For example, if I click Zoom Elite 6, I shouldn't be recommended the Zoom Elite 5...

I have more advice but can't give it out for free.

Helpful responses. Thank you. This was probably the best time of the year to quietly release this, and the worse time if we wanted large number of people to see it amongst all the doping controversy and NCAA/DL/Portland coverage.

In regard to the emphasis on the most reviewed shoes, we purposefully used the language "most popular," not necessarily best. We include the ratings so it's clear that most popular does not necessarily equal "best." Sorting by highest rating is even more absurd, as you get dozens of shoes with high ratings but only a review or 2, which is potentially more misleading. But this is a good point to consider in the language we use. It also may make sense to make a composite that weights ratings by reviews, but composites are often even more confusing.

The suggestion about making it easier to filter discontinued shoes is a good one. We let you filter on the year release date to avoid that issue, but perhaps we should apply that filter by default. A particular problem in that regard is that we need data on the most recent shoes to be useful. It's very difficult to have a lot of data on new shoes as they're new, but data on old shoes is only useful insofar as it informs purchase of newer models and helps people compare new shoes to old shoes. So we'll obviously have to work hard to collect data on NEW shoes (which we've done our best on, frankly, leaving a splash page up to collect data for like 2 weeks). Hopefully with the tool available people will be more apt to submit their experiences.

In general, we still have holes in the data to fill. It's not easy to collect minor details on 30+ shoe companies and 400+ shoes. Especially when a number of these things require judgments (is, for example, the Kinvara a lightweight shoe? A minimal shoe? It's probably lightweight but not minimal. Not everyone will agree.) So that's hard but Weldon has put a ton of sweat into filling out the database. We'll continue working on that in the future.

Every so often I know there are still some software glitches. We're working on improving them. We built the tool using the statistics program R and a framework called Shiny, which lets us us the power of R in real time. Which is awesome. It also let us develop this a lot faster than we would be able to otherwise. But it's still not a perfect framework, and we've even worked directly with the developers in some cases to try to work these bugs out.

Keep the feedback coming.

See what you are doing here, nice first version. Thanks.

It's a lot of work and improving usability will take more effort. I'd like to be able to see several shoes more quickly, side by side or scan a brand instead of having to select each individually. I assume you are heading that way, I think this will be useful.

Maybe you should add kilogrammes and metres in the height and weight fields... I have no bleeding clue what my height in feet and inches is.

FYI: When I clicked on the link to see the detailed review of the Brooks Ravenna I got directed to the Saucony Kinvara review.

Maybe you could have a section/filter/whatever that only shows shoes with sample sizes above the median? Or if there are a lot of 1s and 2s and the median would be like 4, just show shoes with only 30 or more reviews, or something like that. Overall, this is a very cool site.

This was built with Shiny?!? Even cooler. Nice work.

heyyo wrote:

This was built with Shiny?!? Even cooler. Nice work.

Yep, it's all Shiny (and other R packages, of course, like ggplot).

Overall much improved over your first version. As you said your database is the thing to concentrate on now. Once you have the most of the bugs fixed, it will take pushing this beyond the LRC regulars to get the numbers you need.

You'll never have enough data on new shoes to be helpful. As you said all we can do is look at the old version and make inferences. So making only this years shoe the default wouldn't help any, then again a 4 year old version is also fairly useless. Maybe default to the past year or two?

You at least need to link different versions of the same shoe with each other. Simply a section that says "other versions of this shoe" would be enough.

Yeah, the line between categories is extremely blurry (and somewhat pointless IMO), and you'll end up with a ton of shoes listed in multiple categories. Right now it looks like a lot of shoes are not even in a category.. At least need to add a catch-all "daily trainer".

What would be nice instead, if possible, is a slider bar where you can set an upper and lower boundary for weight and Heel/toe drop. Cushioning (based on stack height?) and support would be nice as well where we can select more or less of each, but I realize those two are harder to quantify for a database. That would get rid of the objective classifications and help us find what we are looking for on the continuum. We can determine for ourselves whether they are a lightweight trainer or a racer.

On that topic, it might be beneficial to ask in the review process whether we used the shoe as a racer or a trainer or both.

Keep up the good work!

Well, apparently if you're like me, there are no running shoes that work for you according to the shoe finder.

Neutral, midfoot, medium arch, size 10.5, normal width, road, marathon runner, 80-99 miles a week at 7:15 training pace, 5'10", 155 pounds, Male.

No recommendation.

Gavin Newsome wrote:

Well, apparently if you're like me, there are no running shoes that work for you according to the shoe finder.

Neutral, midfoot, medium arch, size 10.5, normal width, road, marathon runner, 80-99 miles a week at 7:15 training pace, 5'10", 155 pounds, Male.

No recommendation.

Are you being sarcastic? When you have a lot of filters applied there isn't likely to be a perfect match even if there are 10,000 total reviews. Plus, many people did not fully complete the review, just gave a shoe a rating but did not fill out the rest of the survey with the detailed information about themselves. In which case it's not shown in your search if you use even one filter.