[外絮] Kobe:我討厭數據分析

作者: Wojnarowski (@WojVerticalNBA)   2019-08-30 19:05:58
Kobe Bryant explains why he hates analytics in NBA
https://reurl.cc/K6KW7q
Bryant joined ESPN’s TV crew from the US Open on Thursday night to talk about
tennis, which he began playing after retiring from the NBA. He is putting out
a tennis book through his production company called “Legacy and the Queen”t
hat he was promoting.
During his appearance, Bryant was asked by John McEnroe what he thinks of anal
ytics.
Kobe今天去現場觀看美網,也上ESPN和John McEnroe接受採訪,被問到了對於數據分析的
想法。
“I hate it,” Kobe said. “It’s ridiculous. What numbers don’t tell you is
they don’t tell you the emotion. I don’t like analytics.
Kobe:我討厭數據分析。這很荒謬,數字無法告訴你的是,它們無法告訴你球員的情緒。
我從來都不喜歡數據分析。
“You see the numbers, but the numbers don’t tell you how or why they are the
way they are. You have to be able to feel that, to sense that. Tendencies.”
你可以看到數字,但是數字並不會告訴你它們是如何呈現出來的,或者呈現出來的原因。
你必須去感受它們。那叫趨勢。
Bryant then gave an example to illustrate his point. He said a stat could tell
you that a player goes to his left 60 percent of the time, but the stat won’
t tell you the player isn’t really comfortable going to his left and would pr
efer to go to his right. He says that’s the sort of thing you find out by sco
uting and watching video.
Kobe舉了一個例子來說明他的觀點,他說譬如有一項數據可以告訴你一名球員他有60%的
時間會往左側移動,但數據不會告訴你該球員在左側的位置其實不是很舒服且他更喜歡往
右側移動。他說這類東西是要通過偵察和看影片來找到的。
短評或心得:
現役球員裡,PG也發表過類似的觀點
Paul George: ‘I’m not a fan of analytics
https://reurl.cc/5gkxjv
作者: jimmyjou2006 (peaker)   2019-08-30 19:17:00
這段話其實和你喜不喜歡艾佛森關係不大我也認同艾佛森很棒 很多人會用數據誤解他 但這段話不過是說明現在數據不能體現的東西還很多 所以他不喜歡的應該不是數據 而是一堆過度解讀數據的菜迷當然愛結果論就像現在10個球星9個在刷數據,去做對數據體現最高報酬的事,而不是為球隊去做贏球的事,大家都馬去刷命中率助攻籃板
作者: roger2623900 (whitecrow)   2019-08-30 19:25:00
數據好的人出來嘴比較有說服力
作者: harry0508234 (harry0508234)   2019-08-30 21:46:00
雖然不是老大球迷 不過說Clarkson等級是哪來的想法...?
作者: swingingbear (Bear~XD)   2019-08-30 23:42:00
真的 因為他會被瘋狂打臉
作者: baseball1388 (^&^)   2019-08-31 09:43:00
投籃命中率當作打擊率在打
作者: dylancdylanc (davidq)   2019-08-31 22:51:00
龜龜: 我跟你相反 最愛數據了..

Links booklink

Contact Us: admin [ a t ] ucptt.com