The Wall Street Journal - 07.03.2020 - 08.03.2020

A16| Saturday/Sunday, March 7 - 8, 2020 **** THE WALL STREET JOURNAL.

Paris

THE EUROPEAN CYCLING

season began to buckle under

the weight of coronavirus

concerns this week as the

first major race cancelation

hit and several teams tempo-

rarily withdrew from racing.

Organizers officially post-

poned this week’s Strade Bi-

anche races in Northern Italy

on Thursday in compliance

with a government decree

that prohibited all sporting

events held in front of crowds

until April 3. Two other races,

Milan-San Remo and Tirreno-

Adriatico, are also likely to be

canceled this month.

Five teams also announced

that they would skip much of

this spring’s racing.

“It is the responsible thing

to do, for our riders, our team

staff, and the greater public

health,” 25 team doctors from

across the sport wrote in a

letter to cycling authorities

and organizers in Italy and

France this week. “There is

no evidence that the health

environment around these

Italian races is safe.”

EF Pro Cycling and Team

Jumbo-Visma plan to avoid

Italy, while Team Mitchelton-

Scott and Astana are sus-

pending all racing until at

least March 22. Team Ineos,

winner of seven of the past

eight Tours de France, also

withdrew from competition

until late March due to coro-

navirus fears and to mourn

sporting director Nicolas Por-

tal, who died on Tuesday.

They will miss the three Ital-

ian races, along with two in

Belgium and Paris-Nice, which

is due to begin on Sunday.

Cycling is uniquely posi-

tioned among all sports to

suffer the effects of the epi-

demic. The pro circuit is

made up of just 19 men’s

teams and 12 women’s outfits

who spend their seasons com-

peting across Europe along

hundreds of miles of road.

The groundswell in favor

of cancelation began on Tues-

day when the American team,

EF Pro Cycling, wrote to orga-

nizers asking to withdraw

from the three races in Italy.

In a subsequent letter, the

team doctors also made the

case for the races in Italy and

Paris-Nice to be scratched.

They were not only con-

cerned about possible dissem-

ination of the virus, but also

the high burden that bike

races place on local medical

facilities. Crashes are inevita-

ble in this sport and hospitals

stay on high alert.

“There may be inadequate

resources or staffing to care

for our riders, in the midst of

a viral outbreak,” the team

doctors wrote. “Furthermore,

we would be contributing to

the overburdening in such a

situation.

And then, there is the

question of quarantine.

By bouncing around Eu-

rope, teams leave themselves

open to sudden rule changes

that could push them into iso-

lation for weeks at a time.

They need only look at

their rivals on the FDJ-Grou-

pama and Cofidis teams to

see how disruptive it can be.

They are unable to race until

at least mid-March because

many of their staff members

and riders have been confined

to a hotel in the United Arab

Emirates ever since several

cases of coronavirus shut

down the UAE Tour last week.

Deciding When to Challenge a Call

The NBA’s new challenge in behavioral economics: What is the optimal strategy for telling refs they’re wrong?

BYBENCOHEN

SPORTS

Game Time Team Win Probability Added Game Result

POR @ DAL (10/27) :08.4 POR +30.1 Win

OKC @ LAC (11/18) :07.3 LAC +22.9 Win

ATL @ POR (11/10) 1:09 POR +22.9 Win

CLE @ CHI (1/18) :20.1 CHI +14.5 Win

SAC @ SAS (12/6) :15.1 SAS +13.2 Win

Note: Win probability added is the difference between not challenging and

winning a challenge

Source:Inpredictable,WSJreporting

Challenge Accepted

The most successful NBA challenges have all come in the fourth quarter

Organizers postponed this year’s Strade Bianche races in Italy.

LAPRESSE/ FABIO FERRARI/ZUMA PRESS

quarter in a close game is a no-

brainer: challenge. An uncertain

out-of-bounds call in the first min-

ute of the first quarter is another

easy decision: don’t challenge. But

most decisions fall somewhere in

between. They’re judgments about

other people’s judgments. And

coaches reveal their risk tolerance,

probabilistic reasoning and basket-

ball rationality every time they

make this mental calculation.

While the overturn rate is high-

est in the first quarter and lowest

in the fourth quarter, according to

NBA data, those stats don’t really

tell the full story. The best way to

understand the true power of this

new rule is to calculate the swings

in win probabilities after success-

ful challenges—which is precisely

what the Journal did.

No team has benefited more

from the ability to change a call

than Portland. Trail Blazers coach

Terry Stotts is responsible for

three of the league’s 10 most pro-

ductive challenges this season, and

his contributions in those games

were hugely valuable. They added

30.1, 22.9 and 11.7 percentage

Virus Fears Put Brakes on Events

Weather

Shown are today’s noon positions of weather systems and precipitation. Temperature bands are highs for the day.

City Hi Lo W Hi Lo W City Hi LoW Hi LoW

Today Tomorrow Today Tomorrow

City Hi Lo W Hi Lo W

Anchorage 24 21 sn 27 20 sn

Atlanta 57 38 s 62 42 s

Austin 63 47 c 73 60 c

Baltimore 50 29 s 62 41 s

Boise 57 32 sh 55 36 c

Boston 44 30 s 54 42 s

Burlington 32 23 s 45 36 pc

Charlotte 55 30 s 61 37 s

Chicago 53 39 s 65 49 s

Cleveland 40 32 s 61 46 s

Dallas 65 46 pc 68 60 c

Denver 6841pc 6438pc

Detroit 45 33 s 58 45 s

Honolulu 81 69 c 79 66 pc

Houston 65 49 pc 72 59 c

Indianapolis 50 34 s 60 46 s

Kansas City 67 47 pc 69 52 s

Las Vegas 75 54 pc 68 51 pc

Little Rock 61 40 s 68 53 pc

Los Angeles 62 51 c 64 51 pc

Miami 73 62 pc 74 67 pc

Milwaukee 50 40 s 58 47 pc

Minneapolis 55 43 s 63 38 pc

Nashville 55 34 s 66 46 s

New Orleans 64 51 pc 72 59 pc

New York City 45 34 s 58 45 s

Oklahoma City 65 43 s 63 54 c

Omaha 69 46 s 70 45 s

Orlando 66 50 s 73 57 pc

Philadelphia 47 30 s 59 41 s

Phoenix 85 60 pc 74 56 c

Pittsburgh 43 28 s 60 42 s

Portland, Maine 41 25 pc 49 34 pc

Portland, Ore. 49 31 sh 53 30 c

Sacramento 60 40 sh 60 46 c

St. Louis 60 42 s 68 51 s

Salt Lake City 67 45 c 55 42 sh

San Francisco 57 42 c 60 48 c

SantaFe 6338pc 5531c

Seattle 47 33 sh 49 31 c

Sioux Falls 66 47 s 68 30 pc

Wash., D.C. 52 34 s 62 43 s

Amsterdam 50 45 pc 53 42 r

Athens 65 57 pc 64 51 sh

Baghdad 73 51 s 80 53 s

Bangkok 94 79 pc 99 79 pc

Beijing 60 37 pc 54 42 c

Berlin 45 33 sh 50 41 c

Brussels 50 44 pc 51 41 r

Buenos Aires 84 72 s 87 73 pc

Dubai 83 68 pc 84 67 s

Dublin 55 41 r 49 37 sh

Edinburgh 53 42 r 49 39 sh

Frankfurt 48 36 sh 52 41 sh

Geneva 49 31 sh 54 39 sh

Havana 75 64 s 76 62 s

Hong Kong 77 69 pc 77 69 pc

Istanbul 63 53 s 66 52 pc

Jakarta 87 75 c 87 77 sh

Jerusalem 57 44 pc 66 52 s

Johannesburg 79 54 pc 81 56 pc

London 54 45 pc 53 41 sh

Madrid 61 33 pc 65 39 pc

Manila 91 77 pc 93 77 s

Melbourne 69 54 s 71 51 s

Mexico City 77 46 pc 77 50 pc

Milan 61 34 s 58 38 pc

Moscow 53 45 r 52 40 sh

Mumbai 83 71 pc 84 73 pc

Paris 51 45 pc 54 42 r

Rio de Janeiro 80 67 s 81 71 pc

Riyadh 87 57 s 78 54 s

Rome 58 42 t 61 37 s

San Juan 85 73 sh 83 73 sh

Seoul 53 31 c 60 37 s

Shanghai 66 49 pc 61 54 c

Singapore 88 76 t 89 77 sh

Sydney 7265pc 7264pc

Taipei City 79 68 pc 82 67 r

Tokyo 52 45 c 56 49 sh

Toronto 38 30 s 52 42 s

Vancouver 45 31 r 45 30 s

Warsaw 4837sh 4432pc

Zurich 46 29 sh 51 39 c

Today Tomorrow

U.S. Forecasts

International

City Hi LoW Hi LoW

s...sunny; pc... partly cloudy; c...cloudy; sh...showers;

t...t’storms; r...rain; sf...snow flurries; sn...snow; i...ice

Today Tomorrow

Warm

Cold

Stationary

Showers

Rain

T-storms

Snow

Flurries

Ice

<

0s

10s

20s

30s

40s

50s

60s

70s

80s

90s

100 +

Ah g ll

Jacksonville

L e k

CCh l tt

LouL

Pittsburghg

ew Yk

Salt Lakalt LLLLke CCCityyy

Tampa

h

Mempphi

t

City

P Dll

illings

Portlandd

an

ld

Atl t

ew

t

San Diego Ph

LLLosLLsAAAlg

LLL

VVeg

p/l/StPs//.Pa.

.LouLL

ChicCh g

hhington D.C.gton D C

t

Chl

kk rtford

hh

d p

Cleveld

l

Ati

k

Albuquerque

h

Oklahoma CityOklahoma City

an AanAt i

ess

oux Fll

Jack ihg

Chy Phhil d l hih lp

aF

C dd

p

P

h d

liihgh

Tc

Abanyby

Topeka

C bb

AAg t

Ft. Worth

pgd

bil

T t

ip

V Calgaryy

d

Anchorage Honolulu

Jacksonville

Little Rock

Charlotte

Louisville

Pittsburgh

New York

Salt Lake City

Tampa

Nashville

Memphis

Detroit

Kansas

City

El Paso Dallas

Billings

Portland

Miami

San Francisco

Sacramento

Orlando

Atlanta

New Orleans

Houston

San Diego Phoenix

Los Angeles

Las

Vegas

Seattle

Boise

Denver

Mpls./St. Paul

St. Louis

Chicago

Washington D.C.

Boston

Charleston

Milwaukee Hartford

Wichita

Indianapolis

Cleveland

Buffalo

Austin

Helena Bismarck

Albuquerque

Omaha

Oklahoma City

San Antonio

Des Moines

Sioux Falls

JacksonBirmingham

Cheyenne Philadelphia

Reno

Santa Fe

Colorado

Springs

Pierre

Richmond

Raleigh

Tucson

Albany

Topeka

Columbia

Augusta

Ft. Worth

Eugene

Springfield

Mobile

Toronto

Ottawa Montreal

Winnipeg

Vancouver Calgary

Edmonton

60s

70s

40s

20s 30s

10s

0s

-0s

80s

70s

70s

70s

70s

60s

60s

60s

50s

50s

50s

50s

50s

50s

50s

40s

40s

40s

30s

30s

30s

30s

30s

30s

30s

20s

20s

10s 0s 10s

N

BA coaches spend

countless hours craft-

ing game plans, dia-

gramming plays and

managing the person-

alities in their locker rooms. Now

they have something else to worry

about: picking the right time to

twirl their fingers.

This is the first season that

coaches have the power to chal-

lenge one call per game—and only

one call per game—by spinning an

index finger and begging referees

to take another look. They’re not

big fans of the rule change. Nei-

ther are the refs. But even if they

don’t like it, they can take advan-

tage of it.

What’s happening this season is

one of the most interesting, unin-

tentional field experiments in be-

havioral economics. The subjects

are NBA executives, coaches and

players. And the whole thing is

playing out in front of millions of

fans with stakes that academic sci-

entists can only dream about.

Here’s what they’re trying to un-

derstand: In this billion-dollar,

zero-sum industry, what is the op-

timal strategy for telling refs

they’re wrong?

“You can think something was a

good challenge—until there’s a

better challenge,”

said Utah Jazz coach

Quin Snyder. “That’s

the part that some-

times, for lack of a

better word, is chal-

lenging.”

NBA coaches have

challenged roughly

500 calls this season,

and 44% of them have

been overturned after

review. But not all

challenges are equal.

Less than 10% of the

successful challenges

this season added

more than 10 percent-

age points of win

probability, and only

1% increased the win

probability by at least

20 percentage points,

according to The Wall

Street Journal’s anal-

ysis of the 220 over-

turned calls through

last week.

Those statistics would suggest

that an overturned call will dra-

matically change at least one game

in the postseason—which is why

the smartest NBA front offices

pored over data last summer to

understand how to effectively de-

ploy this new weapon. They were

attempting to objectify the subjec-

tive. “So you can make a decision

that makes sense from a probabil-

ity standpoint,” Snyder said.

It was considered such a press-

ing strategic issue that a study

about the optimal time to chal-

lenge was named the winner of

last fall’s NBA Hackathon. The

prize for the two students behind

the paper included lunch with NBA

commissioner Adam Silver.

Once the season began, the real-

ity of a stressful game in a packed

arena and little time to make a de-

cision became even trickier than

coaches anticipated. There are two

questions they have to consider

when they disagree with a call:

What’s the value of winning the

challenge, and what’s the probabil-

ity of winning the challenge?

A phantom foul on a 3-pointer

in the last seconds of the fourth

points of win probability in the

fourth quarter and the last sec-

onds of the third quarter—and the

Blazers won all three games. They

are currently in a heated race for

the No. 8 seed in the playoffs.

Those decisions could be the dif-

ference.

But our analysis showed the

Blazers are outliers. The surprising

thing is that the rule change hasn’t

mattered as much as you might

think. Most successful challenges

have been virtually meaningless.

And some are worse than mean-

ingless. They’re foolish.

There were five successful chal-

lenges of out-of-bounds calls in

the first quarter, for example, but

winning one extra possession that

early in the game is essentially

worthless, and it can actually be

costly if there’s a questionable call

later in the game that can’t be re-

viewed. Even though the coaches

won these challenges, their deci-

sions had about as much effect on

the game as doing the Macarena.

It’s not just out-of-bounds calls.

Many coaches have learned by

now that most first-quarter chal-

lenges are silly. Even though some

prefer to wipe points off the board

whenever they can, there hasn’t

been a single overturned call in

the first quarter worth more than

5 percentage points of win proba-

bility, according to the Journal’s

review. The average swing in win

probability from these decisions

was 2.5 percentage points.

But of the 15 most consequen-

tial overturned calls this season,

12 came in the fourth quarter, two

in the last minute of the third

quarter and one on the last play

before halftime. It might be tempt-

ing to challenge a terrible foul call

in the first quarter or listen to a

player who insists that the refs got

it wrong. It also might be pricey.

The numbers suggest that

coaches prefer keeping a challenge

in their pockets until the fourth

quarter even if that means they

never get a chance to use it. There

had been about 500 challenges

through last week. That may

sound like a lot. It’s not. It means

that coaches didn’t use the chal-

lenge in 70% of their games.

It’s not justwhenthey should

use a challenge that has become a

source of fascination around the

league. It’s alsowhy. The overturn

rate on foul calls is only 38%, but

it’s 69% on goaltending and 76%

on out-of-bounds calls.

None of these statistical in-

sights are novel to the guinea pigs

in the NBA’s latest experiment: G

League coaches. The ability to

challenge one call per game was

implemented in the NBA’s minor

league before the 2017-18 season,

and coaches quickly learned when

to unleash their superpower. “I

never use mine until the mid-

fourth quarter,” said Austin Spurs

coach Blake Ahearn. “That’s when

things can drastically change.”

Golden State Warriors coach

Steve Kerr consulted with the as-

sistants on his staff who had expe-

rience in the G League, and they

were adamant in their recommen-

dation: save the challenge for the

end of a game.

“And I didn’t listen to them,”

Kerr said.

But he’s been among the most

aggressive coaches when it comes

to making refs squint at a court-

side monitor. Entering Friday’s

play, Kerr has challenged 22 calls

this season. The league median is

16. Memphis coach Taylor Jenkins
    has challenged 3. Toronto Raptors
    coach Nick Nurse has challenged
    34—by far the most in the NBA.
    What makes this experiment the
    kind that would make behavioral
    economists drool is that the ex-
    perts appear to be adapting to
    their incentives in real time. Be-
    fore Jan. 1, 23% of challenges were
    in the first half. That number is
    19% after Jan. 1 and 16% since Feb.


17. 
    

    “I hate it,” Atlanta Hawks coach
    Lloyd Pierce said. “But it’s there.
    You have to use it.” ETHAN MILLER/GETTY IMAGES, DOUG MCSCHOOLER/ASSOCIATED PRESS

    
    

This is the first season that NBA

coaches have the power to challenge

one call per game. Toronto Raptors

coach Nick Nurse, below, has

challenged a league-high 34 calls.

BYJOSHUAROBINSON