Field Theory: On Averages
48 Responses
First ←Older Page 1 2 Newer→ Last
-
I can't believe I missed this graph debate. FWIW, I think if you're trying to compare multiple trends on the same graph, then a clustered bar chart would be rubbish.
Technically incorrect use of line graphs is pretty prevalent in my experience of science as well. Even if you have two nominal factors, it's quite quite common to show them as a line graph to present it as a cross-over interaction. (If you [http://www.socialresearchmethods.net/kb/expfact.phpscroll down on this page]] you'll see what I mean).
-
http://www.socialresearchmethods.net/kb/expfact.php
Sorry, my keyboard macron settings destroyed my link.
-
Sorry, this is WAY off topic, but I wonder what y'all think of this...
The Dunedin City Council and the Otago Cricket Association say the concept mirrors "black-outs" created during the All Blacks' games in the city.
Huh? I thought the "black outs" related to the All Blacks name, and the fact they wear black as opposed to the opposition. How does it apply where all sides wear white in test cricket?
I don't get it. I mean, I get the pun. But otherwise, I don't see what the relevance is. Maybe I just haven't had enough coffee this morning. Enough white coffee, anyway.
-
this is WAY off topic, but I wonder what y'all think of this...
Sorry Steve I need to cross-link better, i wrote this on the Dropkicks this morning: Dunedin, they just don't get it
-
Well they clearly haven't thought it through. Not only are there the implications you've discussed, but do we really want anyone to encourage the wearing of white pants?
-
this is WAY off topic, but I wonder what y'all think of this...
It's like our second test match in a gazillion years?
Umm, no, OK, I give up. 'We' do look stupid.
-
Umm, no, OK, I give up. 'We' do look stupid.
Yeah next thing ya know those chumps will dream up some flash new stadium you can't play cricket in. Oh wait.
At least the university oval is great for tests.Ānd I found a white pants diatribe side. I'm sure there are many more.
-
OK James, you've posted three broken links in a morning. Do you need the cheat sheet?
Yeah next thing ya know those chumps will dream up some flash new stadium you can't play cricket in.
I find myself more and more having discussions with people and having to say "oh, I'm in favour of _a_ new stadium. Just not really this one..."
-
Uh, I was fine until I installed Te Taura Whiri's macroniser. I guess I should just try to remember to preview. And yes, I do actually use macrons more frequently than I post links on public address...
I was initially reasonably pro-Stadium, but the more the details are filled in, the dafter it seems.
I can has tohutō!
-
OK James, you've posted three broken links in a morning. Do you need the cheat sheet?
I fixed them.
-
What is also interesting about the Ali Williams stats is that, while his First 3 stats look pretty wild, there appears to be a bit of predictability in his performance - sharp decreases are usually followed by sharp increase and vice versa. Maybe it's a regression-to-the-mean kinda thing.
Until you look at his last three games (albeit missing the Hong Kong match) they are all trending downward
-
Until you look at his last three games (albeit missing the Hong Kong match) they are all trending downward
But [and I'm saying this because the injury thing got me thinking] Tracey's stats don't contain amount of time on the field do they?
That would be a good stat to include I think, so you can get things like 'metres gained per minute'.
-
Ali Williams is erratic and Richie McCaw is awesome.
...or maybe Richie is a #7 with a consistent role and Ali is a tight forward, whose degree of 'tightness' depends on the game plan adopted for a particular opponent. Hell, I played open side flanker for nine years, and in at least two of those years I was the only player on our team who ever made a tackle. In the pre-Michael Jones days, that was pretty much the Job Description right there.
(PS. Can I haz regressions with confidence limits fitted?)
-
But [and I'm saying this because the injury thing got me thinking] Tracey's stats don't contain amount of time on the field do they?
Yes they do. Haydn just hasn't incorporated them.
From the Ireland game from Tracey's page:
An asterisk denotes a player that came on as a substitute. This week: Aofa for Woodcock (yellow card) at 45 min then Woodcock back on for Tialata at 60 min, Weepu for Cowan at 60 min, Toeava for Smith at 63 min, Flynn for Mealamu 65 min, Tialata back on for Woodocock at 67 min, Read for Kaino at 71 min, Donald for Carter at 75 min.
-
Yes they do. Haydn just hasn't incorporated them.
From the Ireland game from Tracey's page:DANGNABBIT!
-
Give a guy a blog and it's all incomplete statistics and line graphs.
Now that Keith is back, maybe he could write a critique and rip it to shreds?
-
Yes they do. Haydn just hasn't incorporated them.
You will notice the piece in the original post where I mentioned not liking data entry. If someone wants to make me a spreadsheet...
Can I haz regressions with confidence limits fitted?
Possibly, though that harder than just generating a bar chart
-
This week: Aofa for Woodcock (yellow card) at 45 min
Also this can't be true because Afoa can't replace Woodcock if he's off for a yellow card. So what really happened was Woodcock came off and then at the first scrum Afoa replaced Kaino and Nonu stepped into the scrum at #8.
-
What is this - the Department of Statistical Rugbyology? I no unnerstand whatchu all talkin' 'bout...
But on the other hand I do loves me a pretty graph! Well done Hadyn.
-
yeah, enough of the chart-bashing. They are obviously indicative stats based on a limited time series. No amount of regression, moving averages, or other fanciful statistical pickiness will give more any meaning to the numbers. They is indicative people! appreciate them for what are... pretty, pretty charts.
-
In the Munster game, does anyone know who was captain after Piri Weepu came off??
-
Mils or Flynn or Read?
-
Post your response…
This topic is closed.