Wednesday, September 05, 2007

USA World Cup Announcer Analysis - 2002 vs 2006

Barring the actual players and teams themselves, there's nothing that stirs up more passion among American soccer fans than the media coverage of our sport. Nowhere does this passion become more apparent than during the FIFA World Cup, the biggest event in the sport. Everything gets put under a microscope, but nothing more so than the announcing.

Jack Edwards and Ty Keough were the commentators for the United States games in 2002, and were routinely bashed by the fans. Edwards was bashed by some in the media due to his unique celebratory phrases, which some deemed too jingoistic. However, that treatment was nothing compared to the reception towards Dave O'Brien and Marcelo Balboa, which was almost uniformly negative. O'Brien had never broadcast a soccer game in his life before 2006, yet had been entrusted to call the sport's top event. Balboa was inexperienced, as well.

I think there's a consensus among American fans concerning the two teams, which is that O'Brien/Balboa were horrible and were a step down from Edwards/Keough. I'm not sure if you could call the Edwards/Keough feelings positive, but they've benefited by the time that's passed and also possibly due to the 2002 team's performance. Nevertheless, they seem to be considered the better of the two pairings. But why exactly is that? Besides the obvious habits that they have, how did the broadcasts differ?

I've taken the time to write out transcripts of two World Cup games so we can find out the answers to those questions. I looked at the openers for both 2002 (USA vs Portugal) and 2006 (USA vs Czech Republic), and in what may be the most monotonous research I've done for this blog, wrote down every word said by both announce teams. It probably took at least 3-4 times the length of the games themselves to get every word correct. I also wasn't able to watch TV or listen to music due to the nature of the work. But that's history, and the end result is that now I can look at all four men statistically to point out their differences, and shed some light on the truth about the feelings we hold toward them. You can download the transcripts here.


One big complaint among soccer fans is that American announcers talk too much, feeling the need to fill up lulls in play with unnecessary banter or stories (Christian Miles does a great job talking about how the differences between soccer and other sports affect announcing here). O'Brien certainly had a reputation for that stemming from his very first match vs Norway in January 2006. It's pretty obvious to point out that Edwards/Keough spent more time calling the game than O'Brien/Balboa, but less time giving background information.

So who actually talked more, and how big of a difference was it? Edwards/Keough said a total of 12,136 words. O'Brien/Balboa said a total of 14,073 words. That's about 16% more. I should note that the Czech game actually was a little over a minute shorter than the Portugal game. Another way of looking at it has the words per second at 2.15 vs 2.50. So yes, O'Brien/Balboa did talk much more frequently.

Now some of you may be thinking that the Portugal game was much more exciting compared to the domination of the Czechs last year. Could that possibly explain the difference? Well, Edwards/Keough did talk more in the second half with their game being close till the end, and O'Brien/Balboa did talk less in their second half. However, on both cases the differences were pretty small, with 51-52% for the more word-filled half. I don't think that had much impact.

What about individual comparisons between the two positions, play by play man and color commentator? In the Edwards/Keough pairing, Jack Edwards spoke 54.65% of the words. In the O'Brien/Balboa pairing, Dave O'Brien spoke 54.70% of the words. So there's virtually no difference there. Both O'Brien and Balboa spoke more than their 2002 counterparts.

It's also very interesting to note that Edwards/Keough had a total of 279 "lines." That is, the number of times they talked before the other person said something. So they had 279, while O'Brien/Balboa had only 202. That means that not only were they talking more, O'Brien/Balboa also went on longer with what they were saying each time they did talk. On average, Edwards/Keough spoke 43.50 words at a time, while O'Brien/Balboa were at 69.67. That's an average of 60% more words each time they spoke, despite 28% fewer lines.

Here's a handy table to recap this info:

Words 12,136 14,073 15.96%
Words per second 2.15 2.52 17.46%

1st half words 47.69% 51.05% 3.36%
2nd half words 52.31% 48.95% -3.36%

PBP man words 54.65% 54.70% 0.05%
Color man words 45.35% 45.30% -0.05%

Total lines 279 202 -27.60%
Words per line 43.50 69.67 60.16%


Which players were mentioned the most? Here's the USA vs Portugal breakdown:

(bold = subbed out)


Friedel 35
Baia 31
Hejduk 51
Rui Jorge 19
Agoos 47
Jorge Costa 23
Pope 54
Couto 32
Sanneh 65
Beto 45
O'Brien 68
Petit 40
Mastroeni 46
Conceicao 52
Beasley 77
Rui Costa 51
Stewart 44
Figo 91
Donovan 70
Joao Pinto 48
McBride 93
Pauleta 39

Jones 43
Paulo Bento 6
Moore 14
Jorge Andrade 5
Llamosa 4
Nuno Gomes 12

Figo obviously the star of the Portuguese team and the guy they focused on the most. Brian McBride with the most mentions of any player, so it's fitting that he would end up with FIFA's man of the match award.

Here's the USA vs Czech Republic breakdown:


Keller 19
Cech 12
Lewis 31
Grygera 19
Onyewu 30
Jankulovski 22
Pope 25
Ujfalusi 8
Cherundolo 19
Rozehnal 10
Convey 37
Galasek 11
Reyna 35
Poborsky 19
Mastroeni 9
Rosicky 43
Beasley 42
Nedved 60
Donovan 40
Plasil 20
McBride 24
Koller 53

O'Brien 23
Lokvenc 13
Johnson 18
Polak 4
Wolff 4
Stajner 5

A very uneven distribution of the opposition players. The top three attacking threats for the Czechs were mentioned many times, but others not so much. Galasek, Rozehnal, and Ujfalusi played all 90 minutes yet were only mentioned a total of 29 times combined. That's less than the lowest total for a 90 minute player in the Portugal game, Vitor Baia with 32. In fact, somewhat maddeningly, 13 of Koller's 53 mentions were in the second half, AFTER he was substituted out of the match.

Overall, this is one of the most striking differences between the two and shows how clearly different the two approaches were. The total number of player mentions in 2002 and 2006 was 1,205 vs 655, which is a difference of 45.6%. This despite the fact that O'Brien/Balboa talked 15% more. Obviously, they had to spend that time talking about other things.

In the table below, the final column shows the percentage of total player mentions among the total number of words. Only last names were counted in this analysis; a handful of times someone may have been called by their first name only (e.g. Landon) but it was very rare.

USA 711 356 -49.93%
Opponent 494 299 -39.47%
Total mentions 1,205 655 -45.64%
USA PCT of total 59.00% 54.35% -4.65%

As a PCT of all words 9.93% 4.65% -5.27%


What terms were favored by each commentary team?

Let's start with something both teams had to deal with equally: they were announcing a USA game. How did they refer to their home country? There are four basic terms that could be used: US, USA, United States, and United States of America. Here's the breakdown:

US 67 64
USA 8 18
United States 67 88
United States of America 1 0

Total 143 170 18.88%

O'Brien/Balboa clearly enjoyed using USA more (or at least O'Brien did). In total, you can see Edwards/Keough referred to the American team 19% less. I guess that can be explained in part by the difference in the total number of words (16%), which that percentage is pretty much in line with. Edwards/Keough also referred to the opposition less, 95 times vs 106 (a 10% difference).

I want to bring special attention to one of the most telling stats for me: the number of times the current score was referenced. I didn't include phrases like "down a goal" or anything similar. Just actual mentions of the current score. The breakdown is as follows:

PBP man 14 22
Color man 1 6

Total 15 28 86.67%

Nearly twice as much for O'Brien/Balboa. This is the type of thing that really annoys. You would think that this is something that would not come up very often, considering that it's on the freaking screen every second of the broadcast.

Here's a look at the frequency of some other terms that they used:

MLS 1 4
Major League Soccer 9 2
FIFA 4 8
World Cup 33 51
Qualifying 4 3
Goal (s) 61 51
Score (s, d) 25 23
Host city's name 6 11
9 2
Group (referring to WC) 1 20
Other two group teams 7 26

Bonus useless info. The 3 letters of the alphabet each team used more than the other (percentage-wise) were as follows: O, S, I for Edwards/Keough, Y, H, E for O'Brien/Balboa. Were player and team names a factor there?


Jack Edwards vs Dave O'Brien

Dave O'Brien is a much better speaker, and is nicer sounding to listen to. However, he showed a clear lack of familiarity with the sport. He failed to really add any analysis to the broadcast, speaking in generalities and sticking to a few key themes (like Koller's height). One thing that I found interesting was that he actually used "proper soccer terminology" more often than Jack Edwards. He clearly does his research of statistics and background information, but that's not a substitute for watching games and commentating on them. Unfortunately, when it comes to using the stuff he's researched, it often adds nothing to the broadcast. Witness the many references to a player's number (14 times), height (7), or age (8).

He really did stick with the themes, just look at the player chart above. For the Czechs, it's Rosicky, Nedved, Koller, and no one else. But it's not just that, or the needless stats. Everything is repeated so many times: Nedved's coming out of retirement, this is the opening game of the World Cup for the USA, the numerous mentions of the score noted above. Possibly because he wasn't used to soccer, he felt the need to keep talking and repeating stuff. There are no commercial breaks in soccer.

Edwards, on the other hand, concentrated almost solely on the game. He always has a sense of urgency in his voice, which made the close game seem all the more exciting. Though, he does get over-excited easily, like whenever the USA scored and famously when the match ended (mine know the rest). That fits into the awkward and/or corny reputation he received, which also includes a tendency to use strange terms frequently.

Whenever he added background info, which was less often than O'Brien, he said it quickly and then got back to the game. Does he talk too much? Possibly, but in comparison it was much less. I do think that Edwards could use a bit more talking about the group and tournament as a whole, and not just call the game like it's in a vacuum. He has a habit of saying just the players last name when he touches the ball, sometimes just calling each touch. O'Brien never did that.

In a way, you can look at this through the previous sporting backgrounds of each man. First and foremost, Edwards is a hockey announcer and O'Brien is a baseball announcer. That may be the most important thing to know here. Just look at the differences between those two sports: Hockey's action is non-stop, with brief stoppages, while baseball takes half a minute after every pitch, with brief moments of action. Hockey is the more closely related game to soccer, which may explain Edwards' superior performance (in my opinion).

I would say most US soccer fans agree that Edwards was better. The catch phrases and the USA results make him more memorable, but that doesn't tell the whole story. Having listened closely to both games I can say with confidence that I prefer Edwards. Whether or not O'Brien has improved since then is another argument.

Ty Keough vs Marcelo Balboa

When it comes to analytical ability, Keough is better. He's not that great, and his analysis is often obvious. He also has an annoying habit of always using a player's full name (which Balboa does sometimes as well, but not nearly as often). But it's hard to find anything good to say about Balboa.

More on him later. Let's delve into Keough's analyses. Keough offered a lot of good comments. He's good at providing perspective on the state of the match, such as where players are lining up and how they've moved around, and provided generalities about the players' skill sets. He also tended to give background info on players, which may have filled a gap in Edwards' call.

I don't like how far too often he was just describing the action on replays. He also cut in way too often despite having little to say. It's very interesting to note that Keough spoke much less in the second half. Throughout the first half he and Edwards were virtually even in words, but in the second half Keough only spoke 42% of the time. I feel that there was a drop off in the quality of his analysis from the first to the second half, possibly because he had to explain how the team shape was early on. The second half featured more obvious stuff, as well as too much talk about "energy" expended.

Balboa...I honestly can't think of one thing I liked about the job he did. I guess he had one or two nice comments about his experience with the US national team. He was canned by ESPN after the tournament and rightly so. First and foremost, he can't talk. So many times he stuttered, mis-spoke, or just phrased something poorly. The worst part was his total disregard for sentence structure and punctuation. Too often he would say multiple things without starting new sentences for each or repeat his point multiple times (see example below). He would also frequently speed up his voice at the wrong times. All of this made for a frustrating listening experience.

He was very predictable as well. The basic pattern went like this:

1. Say something about the action or something that needs to improve
2. Provide a list of 3+ things that happened or need to happen (in one sentence)
3. Repeat the first thing or say what needs to happen again


"Absolutely. Great shot from there. Nobody closing him down, he touches it, just rips one off the crossbar. Kasey Keller couldn't do anything about it, dove up there, couldn't touch it, hit off the crossbar. Let's take a look at, nobody pressuring him, nothing, nobody steps up. Look at that ball knuckling and dipping. Kasey Keller's right there but nobody's putting any pressure. You leave him alone like that, Rosicky alone, he's already punished you once in the thirty-sixth minute. You gotta put some pressure on him."

He was much more opinionated than Keough (which ESPN obviously likes, see Eric Wynalda). The actual analysis wasn't horrible in comparison, worse yes, but the differences there are far overshadowed by the talking problems. Out of 100 lines, 30 began with the word well (usually "well and"). If not well, then some other poor/awkward beginning. I think he had some good analysis in him if he could've said it better. Simply unlistenable.

The difference between Keough and Balboa was much greater than Edwards and O'Brien. Keough is clearly better, and I was actually surprised at the job he did given his reputation among US fans. He was certainly not great, but he worked well with Edwards as a team. He may not stand out as a broadcaster but given the other guys we've listened to, that may be a good thing. Although I'm sure if I watched the Mexico game from 2002 instead, I wouldn't be as positive (going on and on about the handball).

In Closing

Overall, the 2002 team did a much better job of calling the match. That was my belief before this analysis, and it holds true. I'm glad that I now know exactly why I felt that way. The 2002 cup wasn't marketed as much and therefore attracted less of a casual audience (plus it was in the middle of the night). I suppose part of the explanation of both performances was due to ESPN trying to cater to the average sports fan more in 2006.

I was a bit surprised by Keough, better than I remembered. O'Brien also looks a bit better (but not good), only because Balboa was the worst part of that broadcast by far. Edwards/Keough were a better team and focused on the match more. They said less but conveyed more, and they certainly didn't make you want to turn them off specifically because of their commentary.

Comments on "USA World Cup Announcer Analysis - 2002 vs 2006"


Blogger Tim Froh said ... (9:29 PM, September 05, 2007) : 

The most telling statistic for me is the number of times that player names are used. It's a total contrast! It also speaks well to how ESPN wanted to broadcast and package these matches; high on "info" and "opinion," low on substance. Sort of like how sportscenter has changed over the years, no?

Also, great work on this. I still am in awe that you actually sat through these games so many times. I know how tough transcription is, so this is really a Herculean effort. Great stuff.


Anonymous Anonymous said ... (8:47 AM, September 06, 2007) : 

great work with this analysis. it really highlights the differences between the approaches that espn took to the different cups. i think if the commentators could somehow take the best from both styles of commentating, it would be an exemplary combination and would receive the approval of soccer fans.


Anonymous Anonymous said ... (10:11 PM, September 06, 2007) : 

This is amazing. Thanks to your hard work, we now have empirical proof that Dave O'Brien is a terrible soccer announcer, and that the 2006 World Cup was the most trying broadcast for a US soccer fan ever.


Blogger Big Corduroy said ... (2:21 AM, September 07, 2007) : 

Man, this freaking rules, if only for that nice little package of Jack Edwards sound clips you included.

Also, this is incredible analysis.


Blogger arthur said ... (9:13 AM, September 07, 2007) : 

Great analysis. Just wanted to let you know the work and effort you put into this is appreciated.

For as ridiculous as Edwards sounded, I prefer his calls whenever a goal is scored. Much more emphatic and enjoyable than O'Brien's dry "It's in the net!"


Anonymous Anonymous said ... (11:39 AM, September 10, 2007) : 

This is an astounding effort and is really shedding some light on an entirely different topic: EPL commentators, John Motson, Andy Gray (heralded as best in the world) and our own here at home.

The next step when you mentally recover from your task could be to add an England match or an EPL match to the equation.


Blogger Craig said ... (11:41 AM, December 18, 2007) : 

I'm not sure Keough is really all that good in isolation, it's just that Balboa was so bad it makes Keough look good by comparison. I'm by no means sold on O'Brien on pbp, but I think a lot of the shots at him were really collateral damage from Balboa's failings.


Anonymous Anonymous said ... (6:43 PM, January 04, 2009) : 

Interesting how you spend so much time and effort to try and turn something subjective into an objective opinion just so you can pat yourself on the back for allegedly having the correct opinion.

Douchetastic, if I do say so myself.


Blogger Greg Seltzer said ... (1:06 AM, May 28, 2009) : 

So wait... he had a formed opinion about something everyone formed an opinion about, did research on the matter and the research supported his initial opinion.

Now which part of this is the part where he's "douchetastic"?


Anonymous Michael said ... (3:17 AM, July 08, 2009) : 

Great work here. I always hated listening to Balboa. I would be interested to know how other announcing duos would respond to your analysis. What about the Dellacamera/Harkes and Bretos/Sullivan teams? I favor Bretos/Sullivan but think both are better than the two teams you analyzed.


Blogger scaryice said ... (4:40 AM, July 08, 2009) : 

I'll be sure to do this analysis for whomever calls the first USA match at the 2010 World Cup.

I really hope it's a good result, because I have to sit there for hours to transcribe it.

Eventually I'd like to add the 1994 & 1998 openers to this data, and possibly an English broadcast of England's opener as well.


Anonymous Sancho said ... (6:06 AM, December 14, 2009) : 

Just to point that this is a world problem, not only in the US (and not only in soccer). Here, in Brazil, those problems happen as well.


Anonymous John said ... (7:04 PM, October 16, 2012) : 

nice stats


Anonymous said ... (7:18 PM, October 16, 2012) : 

very interesingt statistics, and it is not a big surprise that in the match USA VS Portugal, Luis Figo was the player that was more pronounced since a lot of soccer goes through him, he was the crack in the game


Anonymous fifa world cup 2002 said ... (6:03 PM, November 28, 2012) : 

Such a nice one stats, keep it up dude.


post a comment